Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An analysis of the relationship between injury severity and hospital inpatient costs Butt, Thomas Stephen 1982

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1982_A6_7 B89.pdf [ 6.1MB ]
Metadata
JSON: 831-1.0095314.json
JSON-LD: 831-1.0095314-ld.json
RDF/XML (Pretty): 831-1.0095314-rdf.xml
RDF/JSON: 831-1.0095314-rdf.json
Turtle: 831-1.0095314-turtle.txt
N-Triples: 831-1.0095314-rdf-ntriples.txt
Original Record: 831-1.0095314-source.json
Full Text
831-1.0095314-fulltext.txt
Citation
831-1.0095314.ris

Full Text

AN ANALYSIS OF THE RELATIONSHIP BETWEEN INJURY SEVERITY AND HOSPITAL INPATIENT COSTS by THOMAS STEPHEN BUTT .A., B.Sc, The U n i v e r s i t y of Waterloo, 1972, 1973 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES The Department of Health Care and Epidemiology We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1982 ( c T ) Thomas Stephen Butt, 1982 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 DE-6 <"}/8-n i i ABSTRACT A sample of motor v e h i c l e accident victims h o s p i t a l i z e d at Vancouver General H o s p i t a l , Vancouver, B r i t i s h Columbia, was chosen to analyze the r e l a t i o n s h i p between h o s p i t a l costs and the s e v e r i t y of the patient's i n j u r y or i l l n e s s . Severity was measured using two scales, s p e c i f i c a l l y , the Injury Severity Scale and the Abbreviated Injury Scale. H o s p i t a l costs were also measured, using two d i f f e r e n t methodologies. The f i r s t was the Per Diem costs that were derived by d i v i d i n g a l l r e l a t e d annual costs by the number of patient separation days i n 1975. A Per Diem episodic cost was determined by m u l t i p l y i n g length of stay by the d a i l y average cost. The second approach used a step-down technique that d i s -t r i b u t e d a l l non-patient care r e l a t e d h o s p i t a l services across d i r e c t p a t i e n t care departments, cost centres or programs. Unit costs were then developed f o r each cost centre, depending upon t h e i r annual workload. The 1975 medical record f o r each patient i n the sample was analyzed to determine the number of work units used i n each cost centre during the patient's h o s p i t a l stay. A Step-Down episodic cost was determined by t o t a l l i n g a l l costs from each cost centre that provided services to the p a t i e n t . A paired t-Test did not show a s i g n i f i c a n t d i f f e r e n c e between the Per Diem and Step-Down episodic costs. I t was assumed that the range of s e v e r i t y of the p a t i e n t i n the sample weakened t h i s t-Test, s e v e r i t y measured by I.S.S. was grouped i n low, medium, and high i i i c ategories, or when se v e r i t y was measured by A.I.S., the paired t-Test did show that there was a s i g n i f i c a n t d i f f e r e n c e i n the two costing methodologies. The regression a n a l y s i s i d e n t i f i e d a s i g n i f i c a n t r e l a t i o n s h i p between both episodic costs and s e v e r i t y . The strongest r e l a t i o n s h i p occurred when s e v e r i t y was measured by I.S.S. and costs were determined, using 2 the Step-Down methodology (R = 0 . 2 6 , F = 3 5 . 4 5 ) . When other r e l a t e d independent v a r i a b l e s ( i . e . , death as outcome and operation not per-formed) and a l l i n t e r a c t i o n terms were introduced, the regression c.o-2 e f f i c i e n t increased to R = 0 . 4 5 and the F value increased to F = 2 4 . 9 . Recommendations were made to include a s e v e r i t y r a t i n g on a l l h o s p i t a l i z e d p a t i e n t s ' records to a s s i s t i n patient c l a s s i f i c a t i o n . A f i n a l outcome of this study was i d e n t i f y i n g the value of a Step-Down approach to deter-mining h o s p i t a l costs and i d e n t i f y i n g the l i m i t a t i o n s of the Per Diem methodology of h o s p i t a l accounting. i v TABLE OF CONTENTS Page ABSTRACT i i LIST OF TABLES v i LIST OF EXHIBITS v i i ACKNOWLEDGEMENTS v i i i Chapter I . INTRODUCTION 1 I I . REVIEW OF LITERATURE 8 Web of Causation 8 Economic Evaluation 11 Evaluation Techniques 13 Health Care Valuation 19 Components of Services A f f e c t i n g H o s p i t a l Costs 23 Severity Measurement 29 I I I . METHODOLOGY 34 Patient Data 35 I d e n t i f i c a t i o n of D i r e c t Patient Costs 38 A l l o c a t i o n of Overhead Departments 43 Drugs, Medical/Surgical Supplies, CSR 50 Unit Values 58 Nursing Units 59 Emergency 62 Operating Room 65 Electroencephalography 67 Intensive Care Unit 68 Electrocardiography 69 Laboratory 69 Radiology 70 Pharmacy 72 P h y s i c a l Medicine ' 73 IV. DATA ANALYSIS 77 A. Per Diem versus Step-Down Costing 78 B. Analysis of the Influence of Severity on the V a r i a t i o n i n Cost 89 Test f o r C u r v i l i n e a r i t y 93 Influence of Other Independent Variables 94 Regression Model 105 V TABLE OF CONTENTS (CONT'D) Chapter V. SUMMARY AND CONCLUSIONS Sample Population - MVA Victims 114 Injury Severity Scale - I.S.S. 116 Other Independent Variables 117 Step-Down Costing 117 U t i l i t y of Relationship Between Episodic Cost and Severity 120 APPENDIX A 122 LITERATURE CITED 125 v i LIST OF TABLES Table Page A. MVA Losses and Separations from B r i t i s h Columbia Hospitals 1 B. S o c i e t a l Losses • 15 C. MVA Victims by A.I.S. C l a s s i f i c a t i o n 37 D. Step-Down A l l o c a t i o n 40 E. Paired t-Test Episodic Costs 78 F. Two Factor ANOVA-Episodic Cost and A.I.S 81 G. Two Factor ANOVA-Episodic Cost and I.S.S 82 H. Paired t-Test D a i l y Cost . 84 K. Two Factor ANOVA-Daily Cost and I.S.S 85 L. Two Factor ANOVA-Daily Cost and A.I.S 86 M. Regression Episodic Costs with I.S.S 89 N. Regression Episodic Costs with A.I.S 90 P. Regression of Component Costs with Severity 92 Q. Test f o r C u r v i l i n e a r i t y 93 R. Regression of Episodic Cost with M u l t i p l e Independent Variables and I.S.S. . 97 S. C o - e f f i c i e n t s of Independent Variables and I.S.S. . . 98 T. Regression of Episodic Cost with M u l t i p l e Independent Variables and A.I.S 101 U. C o - e f f i c i e n t s of Independent Variables and A.I.S. . . 102 V. C o - e f f i c i e n t s of Independent Variables i n the Regression Models 107 v i i LIST OF EXHIBITS E x h i b i t s Page P o t e n t i a l Payoff f o r Preventive Program f o r MVA - Figure 1 & IA 4 A l l o c a t i o n Formula "A" Square. Footage of Floor Area 42 A l l o c a t i o n Formula "B" B u i l d i n g Depreciation A l l o c a t i o n Equipment Depreciation 45 A l l o c a t i o n Formula "C" Employee Benefits and S a l a r i e s . 46 A l l o c a t i o n Formula "D" Estimated D i s t r i b u t i o n of Laundry and Linen Expense by Functional Area 49 A l l o c a t i o n Formula "E" Drug Expense From the Cost Centre Printout 51 A l l o c a t i o n Formula "F" S u r g i c a l and Other Medical Supplies 53 A l l o c a t i o n Formula "G" Medical Record Department Services . . . 54 A l l o c a t i o n Formula "H" Nursing Administration 56 Graph 1 Severity by Age 96 Graph 2 - Model 1 Step-Down Cost by Operation, Death and I.S.S 108 Graph 3 - Model 1 Per Diem Cost by Operation, Death and I.S.S 109 Graph 4 - Model 2 Step-Down Cost by Operation, Death, Severity and I n t e r a c t i o n Terms I l l Graph 5 - Model 2 Per Diem Cost by Operation, Death, Severity and I n t e r a c t i o n Terms 112 v i i i ACKNOWLEDGEMENTS This thesis would not have been p o s s i b l e to complete without the unlimited patience and support of a l l those involved i n i t s production. Chris Long deserves p a r t i c u l a r r e c o g n i t i o n for her endurance i n supporting t h i s endeavor. Mary Yates and Agnew Peckham and Associates were of great help i n the production of the many dr a f t copies of t h i s report. Many thanks are given to Bob Evans, who had to supervise t h i s t hesis over several years and across thousands of miles. Because of t h i s distance, valuable assistance was provided l o c a l l y i n the i n t e r p r e t a -t i o n and a n a l y s i s of the research tasks. For t h i s , I am g r a t e f u l to Terry Delmore and Jack Williams of the Health Research Unit, Toronto, and to Jim Harrold, Peter Hawrylyshyn, Evelyn Lazare and Frank Markel. Without t h e i r help, there would not be a completed thesis at t h i s time, or any time i n the future. F i n a l l y , the f i n a n c i a l support provided by I.C.B.C. i s g r a t e f u l l y acknowledged. I t allowed the author time to c o l l e c t the o r i g i n a l data and made poss i b l e in-depth computer a s s i s t e d data a n a l y s i s . INTRODUCTION In the 1974 federal government working document, "A New Perspective on the Health of Canadians," accidental death (to which t r a f f i c death was the largest contributor) was identified as the number one k i l l e r for Canadians below the age of 40. Every year in Canada approximately 6,000 people die i n road accidents and up to 35 times that number are involved 1 2 in injury-producing accidents. A recent Nova Scotia study pointed out that 25% to 30% of a l l motor vehicle accident (MVA) victims are hospital-ized and another 30% use hospital outpatient c l i n i c s . The following table displays separations and patient separation days (i.e. , hospital bed days accumulated by those patients discharged in each calendar year) for MVA victims hospitalized i n Briti s h Columbia for the years 1972 through 1977. MVA victims are identified by the ICD-9 E-codes 810-819, which classify patients by external cause of injury, i.e., motor vehicle t r a f f i c accidents. TABLE A MVA Length of Stay and Separations From Br i t i s h Columbia Hospitals 1972 1973 1974 1975 1976 1977 Separations 12,300 14,500 14,600 13,300 12,800 13,200 Patient Separa-tion Days 124,500 146,200 150,300 134,800 124,800 128,000 Average Length 10.12 10.08 10.29 10.13 9.75 9.69 of Stay Days Days Days Days Days Days Source: Publication of Briti s h Columbia Ministry of Health, Statistics  of Hospitalized Accidents (Annual Issue), Hospital Program Branch, Research Division. ''l.alonde M., A New Perspective of the Health of Canadians: A Working  Document, Ottawa, A p r i l 1974. 2 Chipman M.L., "Hospitalization after MVA i n Nova Scotia," Canadian  Journal of Public Health, Vol.64, March 1973. - 1 -- 2 -The high deaths and i n j u r i e s rates from MVA's are d i s t u r b i n g for two reasons: • the p o t e n t i a l f or preventive programs - various safety and preventive measures could reduce the number of a c c i -d e n t s by countering the e t i o l o g i c a l f a c t ors of MVA's; o the burden on the p u b l i c - the cost of v i c t i m s ' h o s p i t a l i -zation i s imposed on the p u b l i c l y supported health care system. Those who a l l o c a t e s o c i e t y ' s l i m i t e d resources must c o n t i n u a l l y evaluate a s p e c i f i c a l l o c a t i o n of resources r e l a t i v e to other things that could be done with the same resources. T h e r e are two issues to be considered i n t h e ' a l l o c a t i o n of p u b l i c resources to preventive programs f o r motor v e h i c l e accidents. The f i r s t i s whether the returns on the investment necessary to reduce trauma and death through MVA's are better than the returns on investments i n other health programs. What ben e f i t s are l o s t or foregone when resources are applied to motor v e h i c l e accident prevention as opposed to other health care programs such as cancer prevention or reducing the incidence of b i r t h defects? The second issue requires choosing the MVA preventive program which w i l l return the greatest b e n e f i t . Programs to prevent motor v e h i c l e accidents can p o t e n t i a l l y save l i v e s and reduce i n j u r y . In the l a t e 1960's, MVA's were included - 3 -i n a disease c o n t r o l program conducted by the Department of Health, Education and Welfare i n the United States. Despite the seemingly high p o t e n t i a l payoff of some motor v e h i c l e accident programs (see Figure 1), there appears to be considerable uncertainty about t h e i r 3 success. Recommendations c a l l f o r implementation of small educa-t i o n a l or preventive programs with a large emphasis on evaluation of goal attainment f o r use i n future d e c i s i o n s . There are many approaches to d e f i n i n g the consequences of preventive programs, as w i l l be explored i n the next chapter. Of p a r t i c u l a r i n t e r e s t to the author, however, i s the r e l a t i o n s h i p between reducing s e v e r i t y and reducing the cost to the p u b l i c of h o s p i t a l i z i n g v i ctims of motor v e h i c l e accidents. Thus, the main obje c t i v e of t h i s thesis i s to f i n d units of measurement that can adequately define s e v e r i t y and h o s p i t a l costs. I f a s i g n i f i c a n t r e l a t i o n s h i p e x i s t s between s e v e r i t y and h o s p i t a l costs, then c a l c u l a t i n g the impact of a preventive program oriented to only one of these consequences w i l l also determine the impact on the other. H o s p i t a l accounting systems i n Canada do not generate accurate costs f o r component health services provided to i n d i v i d u a l p a t i e n t s . T r a d i t i o n a l methods of cost determination i n a h o s p i t a l s e t t i n g are based on pa t i e n t days as a measure of output; they are much too Grosse R.N., "Cost-Benefit Analysis of Health Services," American  Academy of P o l i t i c a l and S o c i a l Sciences, P h i l . Annals 399, 1972. - 4 -Figure 1 P o t e n t i a l Payoff for Preventive Program for MVA •CANCKK PROGRAMS COMPARED TO OTHER PKOCRAMS 600 500 400 100 110 Horizontal: Deaths Averted—in thousand!, Vertical: Cost in millions of dollars * Includes programs on use of seat belts, defensive driving, and reduction in pedestrian injuries. Figure IA 600 500 -DOLLAR SAVING IN CANCER PROGRAMS COMPARED TO OTHER TREATMENT PROGRAMS-Horizontal: Savings in billions of dollars Vertical: Program costs in millions of dollars * Includes programs on use of seat belts, defensive driving, and reduction in pedestrian injuries. Source: Grosse R.N., "Cost Benefit Analysis of Health Services," American  Academy of P o l i t i c a l and S o c i a l Sciences, P h i l . Annals 399,1972. general to be of a n a l y t i c a l use for s p e c i f i c diagnostic treatment or therapeutic s e r v i c e s . Frequently, programs of care i n a h o s p i t a l s e t t i n g do not n e c e s s a r i l y follow the departmental hierarchy of the organization; thus, the l i n e item budget, the key 4 element i n c o n t r o l of n o t - f o r - p r o f i t organizations , i s not a u s e f u l management c o n t r o l t o o l . Costs are a l l o c a t e d to cost centres, not n e c e s s a r i l y to patient care programs that r e g u l a r l y transcend these groupings. Even with the change to g l o b a l budgets experienced i n Canadian h o s p i t a l s i n the l a s t decade, i d e n t i f i c a t i o n of costs f o r treatment, therapy or diagnostic programs i s s t i l l not r e a d i l y a v a i l a b l e , as there has been no attempt to i d e n t i f y output measures of h o s p i t a l services with t h e i r operational and overhead costs. Just as h o s p i t a l costs are so d i f f i c u l t to i d e n t i f y f o r i n d i v i d u a l patients, so too i s an acceptable code f o r s e v e r i t y . Most cost a n a l y s i s research f a i l s due to an i n a b i l i t y e i t h e r to account f o r the d i f f e r e n t types of patients that h o s p i t a l s treat or to c o n t r o l f o r the i n t e n s i t y of services provided the patient. While patients who have a s p e c i f i c disease may be coded using a t r a d i t i o n a l method of disease c l a s s i f i c a t i o n , i t i s d i f f i c u l t to maintain homogeneity wi t h i n each disease code with respect to h o s p i t a l episodic costs. I t i s the assumption of t h i s author that s e v e r i t y coding would provide e l a s t i c i t y to t r a d i t i o n a l patient c l a s s i f i c a t i o n methods; Decoster D.T. and Schafer E.L., Management Accounting, Wiley and Sons Toronto, 1979, pg.693. - 6 -t h i s could improve cost analysis research. Moreover, se v e r i t y ' s greatest value, p a r t i c u l a r l y i n the analysis of motor v e h i c l e accident v i c t i m s , i s that i t could provide a s i n g l e code f o r m u l t i p l e i n j u r y v i c t i m s . This thesis w i l l deal s o l e l y with the methodology needed to determine, to a reasonable degree of accuracy and f e a s i b i l i t y , the d i r e c t costs of h o s p i t a l services provided to MVA v i c t i m s . Two assessments w i l l be performed to determine i f the chosen methodology r e s u l t s i n a s i g n i f i c a n t d i f f e r e n c e from r e s u l t s of t r a d i t i o n a l Per Diem average costing. F i r s t , episodic costs c a l c u l a t e d by the proposed methodology w i l l be compared to those c a l c u l a t e d by Per Diem average costs; secondly, patient s p e c i f i c d a i l y costs w i l l be compared with d a i l y average of the Per Diem methodology. Data w i l l be analyzed to determine the c o r r e l a t i o n between episodic costs and i n j u r y s e v e r i t y , and between per diem cost and i n j u r y s e v e r i t y . The objective here i s to assess whether s e v e r i t y can be used as a treatment p r o f i l e f o r the determination of h o s p i t a l costs. Such a c o r r e l a t i o n could a s s i s t the decision-making process regarding the implementation of injury-reducing ( i . e . , reducing s e v e r i t y of the impact) preventive measures. With a s i g n i f i c a n t c o r r e l a t i o n , the cost to the h o s p i t a l system could be projected i f the s e v e r i t y l e v e l of i n j u r i e s expected a f t e r implementation of preventive - 7 -programs could be estimated. F i n a l l y , data a n a l y s i s w i l l determine major contributors to episodic cost; i s length of stay the s i g n i f i c a n t i n d i c a t o r , or are other v a r i a b l e s i n the h o s p i t a l treatment process more s i g n i f i c a n t to the determination of t h i s cost? - 8 -I I . REVIEW OF LITERATURE The Oxford Dictionary defines "accident" as follows: "an event happening by chance, without plan or cause; a mishap usually r e s u l t i n g i n harm, i n j u r y , or damage." I f t h i s d e f i n i t i o n i s accurate, and there i s no cause or causal agent f o r an accident, how can counter measures be implemented to l i m i t the e f f e c t s of accidents, e s p e c i a l l y t r a f f i c accidents? Web of Causation Most accidents of our society ( t r a f f i c , home, recreation, and i n d u s t r i a l ) are u s u a l l y personal occurrences varying i n l o c a t i o n and i n degree of s e v e r i t y of outcome. As a r e s u l t , many of these accidents do not get reported and so do not get categorized; i t i s thus d i f f i c u l t to determine a causal agent and then to develop appropriate remedies. What separates t r a f f i c from other types of accidents i s that t h e i r environment i s generally neither stable nor r e p e t i t i v e . T r a f f i c accidents are an e c l e c t i c combination of non-s p e c i f i c events that come together momentarily. The term "accident," i n most cases, i s a misnomer i n an e t i o l o g i c sense. Haddon spoke quite strongly on t h i s concept when he discussed the t r a n s i t i o n to e t i o l o g i c study of trauma.^ He was c r i t i c a l of Haddon W., "The T r a n s i t i o n to Approaches E t i o l o g i c a l l y rather than D e s c r i p t i v e l y Based, with respect to Epidemiology Prevention and Amelioration of Trauma," American Journal of P u b l i c Health, Vol.58 August 1968. - 9 -professionals who brought to t h e i r research personal p r e d i s p o s i t i o n s to such concepts as luck, chance, and mishap; by doing so, they introduced into t h e i r s c i e n t i f i c framework n o n - s c i e n t i f i c elements. 2 In a 1975 report , Haddon furth e r described t h i s r e l a t i o n s h i p with regard to the motor v e h i c l e accident by adding to the normal mechanical energy t r a n s f e r which occurs during a t r a f f i c accident, other concepts such as thermal, e l e c t r i c a l , chemical, and i o n i z a t i o n r a d i a t i o n energy t r a n s f e r s . In addition, he suggested studies that could be made on the human element and made furth e r reference to the growth of research on both the forces that produce i n j u r i e s to 3 animate and inanimate structures, and the ways these can be avoided. F i n a l l y , Haddon stressed the importance of choosing counter measures that reduce the damage, not n e c e s s a r i l y ones that prevent the . . . . r u • 4 i n i t i a t i o n of the impact. Another area of study suggested by Haddon was society's penchant f o r blaming the v i c t i m and ignoring the causal sequence."' This area of the behavioural aspects of the motor v e h i c l e accident i s highly inbred and i s concerned almost e x c l u s i v e l y with the d e s c r i p t i v e , not e t i o l o g i c 2 Haddon W., "Reducing Damage of Motor Ve h i c l e Use," Technology Review, July/August, 1975. 3 Haddon W., Suchman E.A., and K l e i n P., Accident Research - Methods  and Approaches, New York, Harper & Row, 1964. ^Ibid.,3. 5 I b i d . , 3 . - 10 -notion of the t r a f f i c accident. This can be no better i l l u s t r a t e d than the report to the Congress from the Secretary of Transportation, "Alcohol and Highway Safety", w i t h i n which most of the information deals with the e f f e c t a l c ohol has on the person, with l i t t l e or no r e c o g n i t i o n of the causal sequence of the events leading to the accident. For example, the problem i s not only to stop the d r i v e r from drinking, but to also stop the drunk from d r i v i n g . The author has recognized the dearth of information on the r a m i f i c a t i o n s of a l c o h o l and d r i v i n g , and formulated many questions that have to be exhaustively explored before e f f e c t i v e counter measures f o r a l c o h o l r e l a t e d t r a f f i c accidents can be determined. At a recent World Congress on the p r o v i s i o n of medical services to t r a f f i c accident v i c t i m s , f i v e steps were out l i n e d for the management of the motor v e h i c l e accident: • i n v e s t i g a t i o n ; • a n a l y s i s and i n t e r p r e t a t i o n ; • planning c o n t r o l programs; • implementation of counter measures; and • evaluation.^ Schwenger C , A l c o h o l and Highway Safety, A Report to Congress from the Secretary of Transportation, 1967. 7W.H.O., "Accidents i n a S o c i a l Context," Chronicle 27, 1973. - 11 -These f i v e steps support Haddon's e t i o l o g i c a l approach to t r a f f i c accidents, e s p e c i a l l y the planning stage where researchers can work with what i s known, and determine a l t e r n a t i v e s to that which has happened. This leads to the assumption that t r a f f i c accidents can be analyzed by the standard epidemiological methods used to analyze any other disease: a susceptible host, a predisposing environment, an i n c i t i n g agent, and t h e i r i n t e r a c t i o n . Accident data must include information on who had them, where they took place, and when and how they occurred; on d i f f e r e n t classes of i n j u r i e s f o r causal agents; g and on the mechanism of i n j u r y . Economic Evaluation In t h i s era of l i m i t e d resources, health care planners must supply consistent and comparable data, appropriate defence and adequate j u s t i f i c a t i o n of new and ongoing programs. According to Martin, of the f i v e items l i s t e d e a r l i e r , program evaluation i s the ranking 9 issue f o r maintaining an e f f e c t i v e program to manage motor v e h i c l e accidents. Martin divided t h i s evaluation process i n t o two areas, program q u a l i t i e s and program attainment: 0 program q u a l i t i e s — those c h a r a c t e r i s t i c s that d i s t i n g u i s h one program from another, eg. cost, q u a l i t y audit ( s i c ) , appropriateness of management dec i s i o n , a b i l i t y to supply s e r v i c e s Murray J.E., " E t i o l o g y of MVA with S p e c i a l Reference to the Mechanism of Injury," N.E. Journal of Medicine NE 492, June 20, 1968. Q Martin D.L., Health Program Evaluation, A Primer Report to the Federal-P r o v i n c i a l Sub-Committee on Quality of Care and Research and Federal-P r o v i n c i a l Advisory Committee on Health Insurance, 1977 Vol.1. - 12 -• program attainment — measures e f f i c a c y , e f f i c i e n c y , e f f e c t i v e n e s s and side e f f e c t s (unwelcome outcomes of implementing the program). 10 Martin f u r t h e r defined the r e l a t i o n s h i p between program attainment and program q u a l i t y . Nevertheless, only "cost" appears l a t e r i n h i s primer as a j u s t i f i c a t i o n f o r developing more formal techniques i n evaluation of programs, e s p e c i a l l y when "stakes involved i n the d e c i s i o n may be high, incorporating not only programs presently i n operation but also proposed or completed programs when they become precedents f o r future decision."^"'" Martin and h i s working party took the view that a genuine demand e x i s t s by administrators and t h e i r p o l i t i c a l masters f o r greater a c c o u n t a b i l i t y , cost e f f e c t i v e n e s s 12 and cost c o n t r o l i n health care , i . e . f o r program q u a l i t i e s . Developing a comparative r e l a t i o n s h i p between program attainments and program q u a l i t i e s f o r preventive health programs i s the basis of an economic evaluation. Preventive programs are actions made with the expectation of s a t i s f a c t o r i l y achieving objectives which maximize ben e f i t s over costs. Therefore evaluation of preventive programs must be oriented to the future to i d e n t i f y the p o t e n t i a l b e n e f i t s from i n v e s t i n g the costs of implementing preventive programs. Stoddart defines t h i s type of Martin D.L., Health Program Evaluation,A Primer Report to the Federal-P r o v i n c i a l Advisory Committee on Health Insurance, 1977 Vol. I I . op.cit.,9. - 13 -evaluation as economic evaluation — the comparative analysis of a l t e r n a t i v e courses of a c t i o n i n terms of both t h e i r costs and 13 consequences. He further i d e n t i f i e s the value of economic evaluation as i t s a b i l i t y to c l e a r l y i d e n t i f y relevant a l t e r n a t i v e s . The r e a l cost of a preventive program i s not the number of d o l l a r s appearing i n the budget; i t i s the health outcomes or consequences achievable by some other program which have been foregone by committing the resources i n question to the f i r s t program. The two components under study i n t h i s t h e s i s , i.e., s e v e r i t y and h o s p i t a l costs,are both consequences or measures of outcomes of preventive programs. Severity can be grouped with those consequences measuring q u a l i t y of l i f e , while h o s p i t a l treatment costs are e x i s t i n g costs which could be reduced or eliminated i f preventive programs prove e f f e c t i v e . In a majority of economic evaluations, these present and p o t e n t i a l l y avoidable costs have to be distinguished from resource costs required to operate the preventive program!"' Evaluation Techniques A formal and systematic approach to a l l o c a t i n g p u b l i c d o l l a r s , such as economic evaluation, has only been used f o r a generation or so. Stoddart G.L., On Determining the E f f i c i e n c y of Health Programs> Unpublished a r t i c l e , Department of C l i n i c a l Epidemology and B i o s t a t i s t i c s and Department of Economics, McMaster U n i v e r s i t y . l^Stoddart G.L., I b i d . , 13. Klarman H.E., "Applications of Cost Benefit Analysis to the Health Services and the S p e c i a l Case of Technologic Innovation," I n t e r n a t i o n a l  Journal of Health Services, Vol.4, No.2, Nov. 2, 1974. - 14 -As Klarman points out, the e a r l i e s t reviews using these techniques were i n the l a t e 1950's and concentrated on water resources p r o j e c t s . His review of cost b e n e f i t l i t e r a t u r e i d e n t i f i e s most major reports of economic evaluation i n health care up to the mid-1970's. The method of measuring costs and consequences v a r i e s with the subject of evaluation. In most cases, the measurement of resource costs i s s i m i l a r , r e g a r d l e s s of the technique chosen f o r the economic evaluation. For MVA, preventive programs resource costs can be r e a d i l y i d e n t i f i e d . Car manufacturers know the cost of equipping v e h i c l e s with seat b e l t s or a i r bags; s i m i l a r l y , budgets for a d v e r t i s i n g campaigns and p o l i c i n g l e g i s l a t i o n can be e a s i l y determined. The measurement of consequences, though, v a r i e s depending on the question posed i n the evaluation and whether or not the program has been proven e f f e c t i v e . ^ Cost b e n e f i t analysis i s the most extensive technique of economic evaluation. D o l l a r s are the measure commonly 18 used to f a c i l i t a t e comparison of b e n e f i t s across various programs. Thus a l l b e n e f i t s must have a d o l l a r value; these include health se r v i c e s avoided, d i s a b i l i t y days avoided and l i f e years gained. Table B l i s t s Fargan's i n t e r p r e t a t i o n of b e n e f i t s or costs to be 19 avoided through a reduction i n MVA. D i r e c t b e n e f i t s are those Klarman H.E., Ibid.,15. "^Stoddart G.L., o p . c i t . , 13. 18 Musgrave R.A., "Cost Benefit Analysis and the Theory of Public Finance," Journal of Economic L i t e r a t u r e , Vol.7, No.3, 1969. Fargan B.M., "1975 S o c i e t a l Costs of Motor Vehicle Accidents," U.S. Department of Transportation, National Highway T r a f f i c Safety Adminis-t r a t i o n , December 1976, page 30. TABLE B S o c i e t a l Losses Direct ( s i c ) * Production/Consumption market costs Home, Family and Community costs I n d i r e c t ( s i c ) * Medical costs - h o s p i t a l - ph y s i c i a n - coroner - r e h a b i l i t a t i o n V e h i c l e damage Losses to others Legal and Court Funeral Insurance administration Accident i n v e s t i g a t i o n T r a f f i c delay Source: Fargan B.M., "1975 S o c i e t a l Costs of Motor Vehicle Accidents," U.S. Department of Transportation, National Highway T r a f f i c Safety Administration, December 1976, page 30. *The usual usage of i n d i r e c t and d i r e c t costs f o r s o c i e t a l losses are reverse to the d e f i n i t i o n s presented by Fargan. - 16 -re l a t e d to the i n d i v i d u a l that can be evaluated, e.g., p r o d u c t i v i t y ; i n d i r e c t b e n e f i t s are those averted costs f o r services or equipment, e.g., health care. A t h i r d category,called i n t a n g i b l e b e n e f i t s (such as the p o t e n t i a l of the i n d i v i d u a l to provide a volunteer s e r v i c e , or coaching l i t t l e league) does not have a market value and i s not included i n the cost a n a l y s i s . Cost-effectiveness analysis i s another common.form of economic evaluation. This i s an analysis of the effectiveness of seve r a l programs o f f e r i n g comparable outcomes. In these analyses, the consequence i s us u a l l y a s i n g l e e f f e c t , common to a l l programs under review, which may d i f f e r i n magnitude among these various programs. The common e f f e c t i s u s u a l l y measured i n natural u n i t s , e.g., l i f e years gained or d i s a b i l i t y days saved. The consequence i n t h i s type of economic evaluation i s not measured i n d o l l a r s . Key '.components of cost- e f f e c t i v e n e s s studies are the a b i l i t y to give a marginal cost of, e.g., saving a d d i t i o n a l l i f e years gained, and to i d e n t i f y the best mix of programs to provide the maximum l i f e years gained. Since t h i s type of ana l y s i s concentrates on only one consequence, i t loses degrees of s e n s i t i v i t y compared to cost b e n e f i t a n a l y s i s . I t i s , however, most appropriate i n evaluating health care programs where the ob j e c t i v e i s to prolong l i f e . Dittman D.A., and Smith K.R., "Consideration of Benefits and Costs: A Conceptual Framework f o r the Health Planner," Health Care Management  Review, F a l l 1979. - 17 -A c o r o l l a r y to cost - e f f e c t i v e n e s s a n a l y s i s i s c o s t - u t i l i t y a n a l y s i s ; here, the consequence i s measured i n quality-adjusted l i f e years and the costs are adjusted to account f or a l l economic implications of the preventive program. This methodology of economic evaluation i s of d i r e c t importance to t h i s thesis as both v a r i a b l e s being measured, i . e . , s e v e r i t y and h o s p i t a l costs, can be included i n a c o s t - u t i l i t y a n a l y s i s . ^ On the cost side of the equation, program operating costs are modified by: • adding a l l health care costs associated with the side e f f e c t s of implementation of an MVA preventive program, e.g., loss of productive years due to drowning when seat b e l t i n h i b i t e d escape; 0 subtracting a l l savings associated with the programs, e.g., savings i n medical, h o s p i t a l , r e h a b i l i t a t i o n and c u s t o d i a l costs incurred by the v i c t i m ; and 0 f i n a l l y , adding a l l costs associated with t r e a t i n g other diseases f o r v ictims who survived as a r e s u l t of the preventive program, but who would not have survived under the o r i g i n a l conditions. I t i s evident that i d e n t i f i c a t i o n of h o s p i t a l costs i s a v i t a l component of t h i s c o s t - u t i l i t y a n a l y s i s . \ f e i n s t e i n M.C., and Stason W.B., "Foundations of Cost-Effectiveness Analysis f o r Health and Medical P r a c t i c e s , " The New England Journal of  Medicine, March 31, 1977. - 18 -On the other side of the equation, u t i l i t y i s the net h e a l t h -e f f e c t i v e n e s s measured i n quality-adjusted l i f e years saved. This approach f a c i l i t a t e s measurement of the conditions under which the l i f e - y e a r saved i s l i v e d , e.g., a l i f e - y e a r saved i s l e s s rewarding spent i n a chronic h o s p i t a l than spent i n the normal environment of the v i c t i m p r i o r to the accident. The expected number of unadjusted l i f e years i s modified by a f a c t o r that takes into account age, pain 22 and s u f f e r i n g , immobility and l o s t earning of the v i c t i m . I t also allows f o r the v i c t i m ' s subjective comparison between conditions under which saved l i f e years are spent and death. Although methods of economic evaluation that include quality-adjusted l i f e years are 23 r e l a t i v e l y new, there i s l i t t l e doubt of t h e i r value i n r e a l i s t i c a l l y i n t e r p r e t i n g the e f f e c t i v e n e s s of health care programs. Assigning a numerical value to h e a l t h status has become i n c r e a s i n g l y important i n the l a s t s e v e r a l years. Weinstein and Stason, i n a study on the A l l o c a t i o n of Resources to Manage Hypertension, included hea l t h status i n d i c a t o r s as a way to develop quality-adjusted l i f e 24 years. K r i s c h e r reviewed se v e r a l s e v e r i t y scales on the assumption that they were a subset of health status indic e s and that t h e i r Weinstein M.C., and Stason W.B., ibi d . , 2 1 . 23 Torrance G.W., Thomas W.H., and Sackett D.L., "A U t i l i t y Maximization Model for Evaluation of Health Care Programs," Health Services Research, Summer \L972. Weinstein M.C., and Stason W.B., " A l l o c a t i o n of Resources to Manage Hypertension," The New England Journal of Medicine, March 31, 1977. - 19 -properties had value f o r a l l health status i n d i c e s . The i n t e r e s t i n health status i n d i c e s , and e s p e c i a l l y i n s e v e r i t y scales f o r MVA, o r i g i n a t e s i n the need to study the e t i o l o g y of the MVA and to evaluate a l t e r n a t i v e programs, emergency care and long term health 25 care needs. As yet, a s e v e r i t y value of an i n j u r e d MVA v i c t i m has not been used to adjust l i f e q u a l i t y , but there i s merit to developing a r e l a t i o n s h i p between the s e v e r i t y of the i n j u r y as a measure of l i f e q u a l i t y , and the cost to implement the preventive program. Stated another way, can the d o l l a r s invested i n preventive programs reduce s e v e r i t y of MVA victims s i g n i f i c a n t l y and thus improve q u a l i t y of l i f e years saved? This thesis w i l l t r y to i d e n t i f y the s i g n i f i c a n c e of the r e l a t i o n s h i p between i n j u r y s e v e r i t y and h o s p i t a l costs. If there i s s i g n i f i c a n c e , f u r t h e r research on s e v e r i t y could determine i t s value i n i d e n t i f y i n g quality-adjusted l i f e years. Health Care Valuation P l a c i n g a value on the costs and b e n e f i t s can be as d i f f i c u l t as i d e n t i f y i n g which consequences and costs are included i n the a n a l y s i s . I f the component s e r v i c e or item i n each b e n e f i t / c o s t area i s traded i n the market place and bears a p r i c e , that p r i c e can be used. From Table B, only p r o d u c t i v i t y and s o c i e t a l losses are not traded i n the market place. Intangible b e n e f i t s of reducing s o c i e t a l losses can Krischer J.P., "Indexes of Severity: Underlying Concepts," Health Services Research, Summer 1976. - 20 -be eliminated from the cost a n a l y s i s , as mentioned e a r l i e r , but p r o d u c t i v i t y measured as a value of a human l i f e or i n d i s a b i l i t y days must be considered i n a cost b e n e f i t study. Most research has been conducted to date on p l a c i n g a value on p r o d u c t i v i t y and as such t h i s area w i l l not be reviewed i n t h i s t h e s i s . Depending on the assumptions made by researchers, a d o l l a r value can be determined f o r a l l aspects of p r o d u c t i v i t y , e.g., value of human l i f e or d i s a b i l i t y days. I f there i s question as to the accuracy or v a l i d i t y of the market p r i c e of any component i n the cost a n a l y s i s , a resource e x i s t s which allows the development of a shadow p r i c e that estimates an imputed 26 cost. This process i s s l i g h t l y d i f f e r e n t from estimating p r i c e f o r p r o d u c t i v i t y as p r o d u c t i v i t y i s not traded i n the market place and thus does not have an o r i g i n a l p r i c e . Of the many costs i d e n t i f i e d i n Table B, the v a l i d i t y of the health care cost must receive the c l o s e s t s crutiny. Legal fees, f u n e r a l costs and property damage are a l l traded r e g u l a r l y i n the market place and each s e r v i c e has a p r i c e that more or l e s s r e f l e c t s the resources used. The use of h e a l t h resources, e s p e c i a l l y those services provided i n a h o s p i t a l , i s not p r i c e d to accurately r e f l e c t costs. Mushkin.and C o l l i n g s , i n KlarmanH.E., o p . c i t . , 1 5 . - 21 -the l a t e 1950's, emphasized t h i s point when they outlined s e v e r a l 27 ways to estimate expenditures by disease category. While one of t h e i r s o l u t i o n s , using operating costs of s p e c i a l t y f a c i l i t i e s ( i . e . eye h o s p i t a l or T.B. sanitarium) has merit, t h e i r suggestions of using average cost per case or per duration of h o s p i t a l stay s t i l l does not accurately i d e n t i f y the cost of services used by i n d i v i d u a l p a t i e n t s . Further, overhead, t r a i n i n g of health personnel and construction costs were to be a l l o c a t e d on an index of r e l a t i v e importance, but no suggestions f o r developing such an index were given i n t h i s a r t i c l e . Rice was one of the researchers at that time to develop a u s e f u l methodology of estimating cost of i l l n e s s , both d i r e c t and i n d i r e c t costs. Her research included h e a l t h care costs of i l l n e s s e s 28 categorized using the major diagnostic groups of the ICDA code. While her approach i d e n t i f i e d d i f f e r e n t average per diem rates f o r d i f f e r e n t i l l n e s s e s , over a given year (for a l l patients f o r health services under s p e c i f i e d p u b l i c programs or those r e c e i v i n g care i n the p r i v a t e s e c t o r ) , i t was too general to be applied to a s i n g l e h o s p i t a l s e t t i n g or to account f o r m u l t i p l e i n j u r i e s . Mushkin S.J. and C o l l i n g s F.d'A./'Economic Costs of Disease and Injury," P u b l i c Health Reports, Vol.74,No.9, Sept. 1959. 28 Rice D.P., "Estimating The Cost of I l l n e s s , " American Journal of  P u b l i c Health, Vol.57,No.3, March 1967. - 22 -Determining the a c t u a l cost of health services at a h o s p i t a l i s an arduous task. Cost accounting techniques which have proven s u c c e s s f u l i n t r a d i t i o n a l business s e t t i n g s have not been used i n n o t - f o r - p r o f i t h o s p i t a l s i n Canada. According to Holder, many h o s p i t a l managements can not produce a budget that.can compare projected estimates and a c t u a l experience accurately enough to provide c o n t r o l over costs 29 and revenue. But the problems associated with using patient days or any v a r i a t i o n on patient days f o r costing of h o s p i t a l output are w e l l known. Most researchers agree that h o s p i t a l costs should be an aggregate of u n i t costs of i n d i v i d u a l services consumed by the p a t i e n t s . Moreover, breaking down servi c e s , and thus costs i n t o u n i t s , often gives a c l e a r e r p i c t u r e of the v a r i a t i o n s i n cost to perform a s e r v i c e than do t o t a l program f i g u r e s . Fortunately, most programs/ departments/cost centres have established work units that can be costed, as was outlined i n Macdonald and Reuter's report on o b s t e t r i c 30 programs. Current budgeting p r a c t i c e s , even those that incorporate new approaches l i k e Zero Based Budgeting, do not provide information on work unit costs. Thus a separate review of each h o s p i t a l i s needed to i s o l a t e the costs of departmental work u n i t s . The American Holder W.W., "Hospital Budgeting: State of the A r t " , Hospitals and  Health Services Administration, Vol.23 No.2, Spring 1978 30 Macdonald L.K.,. and^Reuter F., "A Patient S p e c i f i c Approach to H o s p i t a l Cost Accounting," Health Services Research, Summer 1973. - 23 -H o s p i t a l A s s o c i a t i o n , i n the l a t e 1950's, developed a basic format f o r the cost determination of work units of a h o s p i t a l . I t i d e n t i f i e d a process to determine d i r e c t and i n d i r e c t patient costs and a basis of a l l o c a t i o n between departments r e l a t e d to the amount of services 31 rendered to each. This process allows i n d i r e c t costs to be a l l o c a t e d to d i r e c t p a t i e n t care to ensure that a l l components providing s e r v i c e to the patient are covered by the u n i t cost of that patient s e r v i c e / 32 work u n i t . Evans and Robinson r e f i n e d t h i s technique by developing simultaneous equations that allow f o r the a l l o c a t i o n of services from departments that serve each other. This i s . t h e methodology that w i l l be used i n t h i s report to i d e n t i f y the costs of the work units of each d i r e c t patient s e r v i c e i n the study h o s p i t a l , Vancouver General H o s p i t a l . Components of Services A f f e c t i n g H o s p i t a l Costs The development of preventive programs f o r MVA victims to reduce health care costs could be f u r t h e r r e f i n e d to focus on those components, e.g. , types of i n j u r i e s , number of patients, or demographics of patients which have the highest influence on cost. Several attempts have been made to standardize an approach to h o s p i t a l costing by adjusting f o r differences 33 i n the number and types of services a v a i l a b l e or by t r y i n g to incorporate TT American H o s p i t a l A s s o c i a t i o n , Cost Findings f o r H o s p i t a l s , Chicago, 1957. "^^Evans R.G., and Robinson G.C., Evaluation of the Economic Implications of a Day Care Surgery Unit, Report to Children's H o s p i t a l , Vancouver, B.C., 1973. 33 Berry R.E., "Product Heterogenity and H o s p i t a l Cost A n a l y s i s , " Inquiry, 7-67, March 1970. - 24 -h o s p i t a l s i z e , u t i l i z a t i o n and an i n f l a t i o n a r y f a c t o r into d e f i n i t i o n 34 of h o s p i t a l costs over time. Frank used a variance analysis methodology to develop an approach to measure comparability between h o s p i t a l s on a t o t a l cost basis f o r a l l U.S. non-governmental, n o t - f o r - p r o f i t , short-35 term, general h o s p i t a l s from 1950 to 1973. These approaches give an understanding of the influences on t o t a l h o s p i t a l costs, e s p e c i a l l y over time, but provide l i t t l e d e f i n i t i o n as to the a c t u a l cost of s p e c i f i c p a tient r e l a t e d s e r v i c e s . A methodology i s needed that r e l a t e s an appropriate accounting technology to the various component services offered i n a h o s p i t a l s e t t i n g . To ensure resource and cost c o n t r o l , a fundamental understanding i s needed of how to define, i n a manageable fashion, the services provided by h o s p i t a l s . T r a d i t i o n a l l y cost c o n t r o l has been succ e s s f u l i n settings where well-defined s e r v i c e s , with a p r e d i c t a b l e set of associated costs, 36 are provided. This begs the question of how to measure changes i n h o s p i t a l cost. Thompson stated that i f patients could be c l a s s i f i e d using c l i n i c a l and demographic v a r i a b l e s , there could be developed a d i r e c t r e l a t i o n s h i p between case mix of the h o s p i t a l and resources used and costs i n c u r r e d . ^ Lave J.R., and Lave L.B., "Hospital Cost Functions," American Economic  Review, 60:379, June 1970. 35 Frank W.G., "A Managerial Accounting A n a l y s i s of H o s p i t a l Costs," Health Services Research, Spring 1976. 36 Thompson J.D., "On Reasonable Cost of H o s p i t a l S e r v i c e s , " Millbank  Memorial Fund Quarterly, 46(1):33 January 1968, Part 2. 37 Thompson J.D., A v e r i l l R.F., and F e t t e r R.B., "Planning Budgeting and Controlling-One Look at the Future: Case Mix Cost Accounting," Health  Services Research, Summer 1979. . - 25 -Disease costing i s one approach to c l a s s i f i c a t i o n of patients; i t groups a l l i n p a t i e n t s with l i k e diseases or h o s p i t a l procedures, e.g., 38 hysterectomy, g a l l bladder surgery, et cetera, but because disease costing r e f l e c t s an average of the costs i n one h o s p i t a l s e t t i n g f o r a l l patients with l i k e diseases i n one time period, the consequence of providing care to i n d i v i d u a l patients i s not apparent, e.g., the i n d i v i d u a l s e v e r i t y of the patient i s l o s t due to the averaging of a l l those i n the disease group. Comparisons of disease costing r e s u l t s between h o s p i t a l centres assume that, the patient populations are s i m i l a r , and that the comparison of the disease cost i s , i n f a c t , a d i r e c t measure of the q u a l i t y of care; t h i s assumption i s not n e c e s s a r i l y correct. Another more soph i s t i c a t e d patient c l a s s i f i c a t i o n system, the Diagnostic Related Groups (DRG) system, was developed at the Yale U n i v e r s i t y Center f o r Health Studies. The DRG approach i s more 39 meaningful than disease costing f o r s e v e r a l reasons: • i t has a manageable number of patien': classes—383 DRG's; • each DRG i s medically meaningful; and » the DRG's demonstrate a s t a t i s t i c a l l y stable d i s t r i b u t i o n of resource uses w i t h i n the universe of patients treated by the h o s p i t a l . Thus the DRG's provide a means of e s t a b l i s h i n g a .hospital case mix by determining the r e l a t i v e number of patients discharged w i t h i n each diagnostic r e l a t e d group. 38 "Lay C , "What i s Disease Costing?" Unpublished Report, Health Admin-i s t r a t i o n , Faculty of Administration, U n i v e r s i t y of Ottawa, February 1974. 39 Thompson et a l . , o p . c i t . , 37. - 26 -Breaking down a h o s p i t a l ' s operating budget into diagnostic groups approximates analyzing h o s p i t a l costs. I d e n t i f y i n g s h i f t s i n u t i l i z a t i o n and corresponding budgetary changes can give i n s i g h t i n both the i n t e n s i t y of care and the fundamental trends i n the nature of the m e d i c a l / s u r g i c a l p r a c t i c e of the h o s p i t a l . ^ This l a s t issue may be more pertinent to non-Canadian h o s p i t a l s at t h i s time, as there i s l i t t l e i n c e n t i v e to save b u i l t i n t o most Canadian h o s p i t a l budgets. As long as these budgets are determined based on h i s t o r i c u t i l i z a t i o n , any help i n making pro j e c t i o n s which are accurate enough to provide managerial controls over costs and reimbursements would be of great 41 assistance to management. While Diagnostic Related Groupings may provide better d e t a i l than disease costing and may r e l a t e better to data bases as they are components of the ICDA-9 coding system, they s t i l l do not f a c i l i t a t e costing i n d i v i d u a l patient care. Nor do they n e c e s s a r i l y allow i n f o r -mation on patients who have more than one disease i n d i f f e r e n t 42 diagnostic r e l a t e d groups. Lynch found that the data base could supply p a t i e n t - r e l a t e d information on more than one DRG but, using patient abstract information, could not i d e n t i f y a weighting system to rank which group had the most impact on the patient's use of h o s p i t a l resources. —-Bowman R., "DRG's Help Trace Revenue Sources," Modern Health Care, January 1980. 41 Thompson et a l . , o p . c i t . , 3 7 . Lynch T., "Report to the Information Systems Steering Committee of Ontario M i n i s t r y of Health," Ontario H o s p i t a l A s s o c i a t i o n , October 20, 1978. - 27 -Others have t r i e d to develop l e s s demanding surrogates f o r case mix by analyzing other c h a r a c t e r i s t i c s l e s s d e t a i l e d than patient c l a s s i f i c a t i o n and have found that " s i z e , number of f a c i l i t i e s and s e r v i c e s , and teaching status of a h o s p i t a l " are c o r r e l a t e d to case mix but could explain too small a portion of the v a r i a t i o n to be A3 of s a t i s f a c t o r y use. F e l d s t e i n used i n p a t i e n t components of a h o s p i t a l , e.g., medical and s u r g i c a l care, for a n a l y s i s of cost . . 44 v a r i a t i o n . 45 J e f f e r s and Siebert f e l t that none of these approaches alone would be b e n e f i c i a l i n explaining v a r i a t i o n i n costs, e s p e c i a l l y among two or more h o s p i t a l s . They added se r v i c e i n t e n s i t y ( E l n i c k i defined t h i s as a measure of the q u a n t i t i e s or volumes of outputs from various departments consumed e i t h e r d i r e c t l y or i n d i r e c t l y by the patient/ 46 physician) and input p r o d u c t i v i t y f a c t o r s to case mix to further i d e n t i f y the v a r i a b l e s that a f f e c t h o s p i t a l cost. They found that gross s e r v i c e i n t e n s i t y accounted for as great a proportion of the increases i n cost per case as did using input p r i c e s , and they concluded that s e r v i c e i n t e n s i t y was a major influence a f f e c t i n g the upgrading of tech-nology and o v e r a l l attempts to improving q u a l i t y and patient care. A 3 Lave J.R., and Lave L.B., "The Extent of Role D i f f e r e n t i a t i o n Among Ho s p i t a l s , " Health Services Research, Spring 1971. ^ F e l d s t e i n M., "Hospital Cost V a r i a t i o n and Case Mix D i f f e r e n c e s , " Medical Care, 3, 95-103, April-June 1965. ^ J e f f e r s J.R. , and Siebert CD., "Measurement of H o s p i t a l Cost V a r i a t i o n . Case Mix Service I n t e n s i t y and Input P r o d u c t i v i t y Factors," Health  Services Research, Winter 1974. 46 E l n i c k i R.A., " H o s p i t a l P r o d u c t i v i t y , Service I n t e n s i t y and Cost," Health Services Research, Winter 1974. - 28 -Combining service i n t e n s i t y ( i . e . , a l l o c a t i o n of h o s p i t a l resources) and a measure of the s e v e r i t y of the patient's i l l n e s s or i n j u r y should pro-duce a s a t i s f a c t o r y approach to measuring h o s p i t a l costs. Severity, as a c l a s s i f i c a t i o n of case mix, allows: a a simple method of grouping patients who may be anatomically d i f f e r e n t (and thus c l a s s i f i e d i n d i f f e r e n t DRG's) but have a comparable s e v e r i t y of i l l n e s s or i n j u r y and • allows f o r patients with multiple i l l n e s s e s who are d i f f i c u l t , to group (as noted e a r l i e r i n the review of Rice's research). - 29 -Severity Measurement Throughout most of the l i t e r a t u r e , a v a i l a b i l i t y of data i s a c r i t i c a l i s sue. The cost of access to data, with the appropriate l e v e l of d e t a i l , e.g., i n d i v i d u a l p a tient surveys, may outweigh the q u a l i t y of r e s u l t s . Information can be taken d i r e c t l y from patient charts, e s p e c i a l l y now that so much information i s stored using disease c l a s s i f i c a t i o n systems. Two approaches to measuring outcomes of surgery were attempted by the s t a f f of Stanford Center f o r Health 47 Care Research: one was an inte n s i v e study with trained personnel interviewing patient, surgeon, anaesthetist, and nurse to c o l l e c t p a t i e n t c l a s s i f i c a t i o n data; the p a r a l l e l study used summary data from P.A.S. Despite the differences i n source data, the r e s u l t s were s i m i l a r . The s t a f f concluded that h o s p i t a l abstract data can 48 be used to measure the q u a l i t y of s u r g i c a l care i n h o s p i t a l s . While such research i s b e n e f i c i a l f o r measuring q u a l i t y of care and f o r showing that abstract information has u s e f u l p r e d i c t i v e powers, i t s c o n t r i b u t i o n to case mix i s l i m i t e d to ranking h o s p i t a l s of s i m i l a r q u a l i t y . I t does not show a matching between work units of the h o s p i t a l and patient c l a s s i f i c a t i o n . What i s needed to balance q u a l i t y of outcome i s a qu a n t i t a t i v e evaluation of patients coming int o the system; t h i s would a s s i s t i n the a l l o c a t i o n of resources and would permit comparisons of h o s p i t a l s i n t h e i r use or a l l o c a t i o n of resources. Stanford Center f o r Health Care Research, "Comparison of Hospitals with Regard to Outcomes of Surgery," Health Services Research, Summer 1976. f^ Stanford Center f o r Health Care Research, Ibid.,47 - 30 -The ub i q u i t y of statements emphasizing the need f o r refinement of 49 50 emergency h o s p i t a l records f o r MVA victims ' reveals the inade-quate state of data c o l l e c t i o n and data development f o r MVA. In these times of co n s t r a i n t i t w i l l be only through better u t i l i z a t i o n of pres-e n t l y a v a i l a b l e resources that any improvements can be e f f e c t i v e i n the organization of the h e a l t h / h o s p i t a l services to MVA and other emergency p a t i e n t s , commonly c a l l e d the Emergency Medical System (EMS). Evaluation of such a system or for that matter any large o r g a n i z a t i o n a l s t r u c t u r e , requires q u a n t i t a t i v e tools which support a n a l y t i c a l approaches such as computer simulation and modelling. As w e l l , since the EMS i s composed of several subsections ( i . e . , the automobile crash; on s i t e treatment; ambulance s e r v i c e ; emergency department services; acute care treatment and post-acute care treatment) there i s a need f o r input and output d e s c r i p t o r s f o r each subsection to monitor a l l EMS programs. Thus any change i n the EMS (e.g., i n t r o d u c t i o n of a preventive program 52 to reduce the second c o l l i s i o n ) oriented to reducing v i c t i m s ' p o t e n t i a l for i n j u r y , must be analyzed by i t s influence across a l l subsections of the EMS. Severity can meet both of these c r i t e r i a as a r e s u l t of recent attempts to quantify t h i s v a r i a b l e . 49 . . . . Noble J . et al., Emergency Medical Services: Behavioural Planning Perspectives, New York, Behavioural P u b l i c a t i o n s , 1973. "^Boyd D.R., Lowe R.J. and Baker R.J., "Trauma Registry New Computer Method f or M u l t i f a c t o r i a l Evaluation of Major Health Problems," JAMA Vol. 223:422, 22 Jan. 1973. "^Cowley R.A., Hudson F. and Scanlon E., "A Prognostic Index f o r Severe Trauma," Journal of Trauma, December 1974. 52 Haddon W., op . c i t . , 2 . - 31 -53 Baker suggests that the se v e r i t y of the i n j u r y , which correlates with the l i f e - t h r e a t e n i n g p o t e n t i a l of the i n j u r y , i s the l o g i c a l descriptor to define and analyze the various subsections of the emergency medical system. Ogawa and Sugimoto also agree that s e v e r i t y i s a good descriptor 54 of patients, e s p e c i a l l y those using the EMS. They reviewed the p o t e n t i a l of incorporating a s e v e r i t y r a t i n g s e r v i c e within the EMS for ambulance attendants that would a s s i s t i n d e l i v e r i n g the appropriate care to the v i c t i m i n the minimum amount of time. E a r l i e r i t was suggested that s e v e r i t y would be an excel l e n t modifier to case mix i n anal y s i s of varying h o s p i t a l costs. Combining patients i n t o groups on the basis of se v e r i t y of i n j u r y requires the use of such scales as Abbreviated Injury Scale^^(AIS) and the 56 Comprehensive Injury Scale (CRIS). These scales were developed to provide a method f o r r a t i n g and comparing i n j u r i e s incurred i n automobile crashes. Both p e r t a i n to i n d i v i d u a l i n j u r i e s but the AIS i s the simpler of the two and thus more widely used. The body i s divided i n t o s i x anatomical categories, s p e c i f i c a l l y head or neck, face, chest, abdominal or p e l v i c contents, extremities or p e l v i c g i r d l e and general. I n j u r i e s 53 Baker S.P., O ' N e i l l B. and Haddon W., "The Injury Severity Score - A Method f or Describing Patients with M u l t i p l e I n j u r i e s and Evaluating Emergency care," Journal of Trauma, March 1974. 54 Ogawa M. and Sugimoto T., "Rating Severity of the Injured by Ambulance Attendant; F i e l d Research of Trauma Index," Journal of Trauma, Nov. 1974. "^Committee on Medical Aspects of Automotive Safety, "Rating the Severity of Tissue Damage I The Abbreviated Injury Scale," JAMA, January 11, 1971 Vol.215, No.2. "^Committee on Medical Aspects of Automotive Safety, "Rating the Severity of Tissue Damage II The Con; rehensive Injury Scale," JAMA, May 1, 1972, Vol.220, No.5. - 32 -i n each category were categorized using the same numerical ranking system f o r s e v e r i t y (1 = minor, 2 = moderate, 3 = severe, not l i f e threatening, 4 = severe l i f e threatening s u r v i v a l probable, 5 = c r i t i c a l , s u r v i v a l uncertain). The A.I.S. i s of value for measuring s e v e r i t y of victims with i n j u r i e s i n only one of the s i x anatomical categories. Most h o s p i t a l i z e d victims of MVA have more than one i n j u r y , with the average being 1.6 i n j u r i e s per patient. 5 7 The authors of the A.I.S. caution against averaging A.I.S. r a t i n g s for each body region as "the q u a n t i t a t i v e r e l a t i o n s h i p of the A.I.S. code i s not known and i s almost c e r t a i n l y ..58 non-linear. Baker et a l . used a quadratic r e l a t i o n s h i p between the top three A.I.S. scores ( i . e . , squared the highest s e v e r i t y scale w i t h i n each body category and i n a d d i t i o n the top three scores from d i f f e r e n t body areas). They l i s t e d t h i s new score f o r patients with m u l t i p l e i n j u r i e s against 59 m o r t a l i t y rates and found s i g n i f i c a n t c o r r e l a t i o n . This "Injury Severity Scale" was defined as the sum ^of the squares of the highest A.I.S. grade i n each of the three most severely injured body a r e a s . ^ "Moylan J.A., Detmer D.E., and Rose J.."Evaluation of the Quality of Hospital Care f o r Major Trauma," Journal of Trauma, Vol.16 #7, J u l y 1976. 58 "Baker S.P., O ' N e i l l B., Haddon W., and Long W.B., "The Injury Severity Scale: Development and P o t e n t i a l Usefulness" Proceedings of the 18th Conference of the American A s s o c i a t i o n f o r Automobile Medicine. Lake B l u f f , 111. AAAM 1974.. 59 Baker et a l . , i b i d . , 58. 6 0Baker et al.» i b i d . , 58. -33-Semmlow"'1' compared h i s r e s u l t s with those of Baker and demonstrated remarkable s i m i l a r i t y despite the complete independence of the data. He also showed that I.S.S. measurements had a strong l i n e a r r e l a t i o n -ship with intermediate EMS system v a r i a b l e s such as length of stay and need f o r s u r g i c a l procedures, e s p e c i a l l y at the lower end of the I.S.S. Semmlow J.L. and Cone R., " U t i l i t y of the Injury Severity Score: A Confirmation," Health Services Research, Spring 1976. - 34 -II I . METHODOLOGY Before any r e l a t i o n s h i p between s e v e r i t y and cost can be established, the s e r v i c e s a v a i l a b l e at the h o s p i t a l and those used by the MVA vic t i m s must be i d e n t i f i e d . As t h i s paper i s an exercise i n developing a methodology regarding cost of h o s p i t a l services for MVA v i c t i m s , only one h o s p i t a l was used to determine the value of t h i s methodology. Vancouver General was selected: i t s large emergency department i n d i c a t e s a s u b s t a n t i a l population of MVA vi c t i m s . To t e s t other hypotheses regarding i n t e r - h o s p i t a l r e l a t i o n s h i p s of se v e r i t y and cost, the methodology i s tr a n s f e r a b l e to any h o s p i t a l that: • l i s t s separation information of patients as i s done on the B.C.H.P. separation tapes; and • produces yearly H.S.I and H.S.2 forms f o r the f e d e r a l government. The B.C.H.P. separation tapes are c r i t i c a l to the ana l y s i s of patient data, as they supply the patient's admission number (thus access to the medical record number) and other pertinent information r e s u l t i n g from the patient's stay i n the h o s p i t a l , such as type of i l l n e s s or i n j u r y and cause of i n j u r y . H.S.I and H.S.2 forms give a l l important a l l o c a t i o n and budget f i g u r e s f o r the major cost centres of the h o s p i t a l . - 35 -Patient Data C r i t i c a l information on the B.C.H.P. separation tapes was the I.CD.A. coding of the cause of the accident (the E code) and the nature of the i n j u r y (the N code). From the E code, the MVA v i c t i m could be s i n g l e d out from the remainder of the patient load of the h o s p i t a l during any one year; with the N code, s e v e r i t y of the primary i n j u r y could be determined on the A.I.S. scale, using Baker and Faigan's mapping of A.I.S. values onto the I.CD.A. codes. ^  This would i d e n t i f y which l e v e l s of s e v e r i t y were represented i n the population; the sampling methodology would ensure representation from a l l l e v e l s . The year 1975 was the l a t e s t year f o r which both.B.C.H.P. separation tapes and the H.S.I and H.S.2 forms were a v a i l a b l e at the time commencing t h i s research at Vancouver General H o s p i t a l . For that year, Vancouver General H o s p i t a l had 906 (out of over 50,000 patients) i d e n t i f i e d as being MVA v i c t i m s . These included persons involved i n MVA while d r i v i n g or being driven i n a car on a p u b l i c road, and pedestrians or b i c y c l i s t s who were struck by or did s t r i k e ( i n the case of b i c y c l i s t s ) motor v e h i c l e s . This f i g u r e of 906 does not represent a l l MVA victims treated at Vancouver General H o s p i t a l i n 1975, j u s t those who were admitted to the h o s p i t a l during 1975. As t h i s study focusses p r i m a r i l y on the i n p a t i e n t cost patterns of MVA v i c t i m s , i t was f e l t that a s i z e a b l e sample of these 906 victims would Personal communications with Barbara Faigan, June 1977. - 36 -be needed to give a s i g n i f i c a n t i n d i c a t i o n of the r e l a t i o n s h i p between s e v e r i t y and the episodic h o s p i t a l costs. As mentioned above, an A.I.S. value could be given to each patient by the B.C.H.P. tapes; but since these tapes do not denote the top three i n j u r i e s and loc a t i o n s , i t was impossible to a f f i x an I.S.S. value on any of these 906 v i c t i m s at t h i s stage of the research. A f t e r mapping of the A.I.S. values with the i n j u r i e s of the 906 v i c t i m s , i t was possible to s t r a t i f y the population to ensure general representation of a l l s e v e r i t y l e v e l s i n the sample. Considering time constraints and the demands involved i n data c o l l e c t i o n from medical records, a 15% sample of the 906 victims was f e l t to be adequate. Consideration was given to taking 15% of each of the ten s t r a t i f i e d groups i n the table below, but t h i s would not give adequate representation from a l l ten s t r a t a because large numbers i n both male and female would appear i n Level 3, but small representation would be made i n a l l other s t r a t a . With these low numbers, a 15% cut-off would be of l i t t l e value. Therefore, to ensure good representation of these other groups, a d d i t i o n a l members from each i n d i v i d u a l stratum were chosen randomly to give the following breakdown among the ten s t r a t a f o r the study sample. - 37 -TABLE C MVA Victims By A.I.S. C l a s s i f i c a t i o n A.I.S. T o t a l 15% Study Sample Severity L e v e l Population Sample Male Female Male Female Male Female 1 21 6 3 1 13 6 2 94 49 14 7 23 16 3 397 213 60 31 60 30 4 58 39 8 5 17 14 5 19 10 3 1 13 . 10 589 317 88 45 126 76 Because the o b j e c t i v e of t h i s study was to determine episodic cost, the sampling technique of random s e l e c t i n g from s t r a t a breakdown based on sex and s e v e r i t y did not n e c e s s a r i l y influence the cost f a c t o r . An increase i n the marginal areas of s e v e r i t y should ensure b e t t e r representation of the whole population than a 15% sample from across each stratum or across the whole population. Once the sample had been selected, each patient's record underwent a complete s e v e r i t y review using both A.I.S. and I.S.S. to determine f i r s t h a n d the s e v e r i t y c l a s s i f i c a t i o n of the p a t i e n t . Therefore, the mapping of the A.I.S. to I.CD.A. codes provided an e x c e l l e n t method f o r sample s e l e c t i o n ; - 38 -but the mapping did not i d e n t i f y the patient's s e v e r i t y index used i n the evaluation process of t h i s t h e s i s . The s e v e r i t y index was c a l c u l a t e d by the author from the p a t i e n t s ' medical records. For the 202 sample MVA v i c t i m s , each medical record was examined to determine the extent of services o r i g i n a t i n g i n each of the d i r e c t patient departments or cost centres. These include nursing u n i t s , time i n the OR, types of procedures i n radiology, and laboratory t e s t s . I d e n t i f y i n g t h i s workload f o r each d i r e c t patient s e r v i c e was the f i r s t step i n the i d e n t i f i c a t i o n of the episodic costs of each patien t . This could only be done a f t e r a u n i t p r i c e had been established f o r the work unit of each d i r e c t patient s e r v i c e . I d e n t i f i c a t i o n of D i r e c t Patient Costs To s t r i c t l y i d e n t i f y the t o t a l d o l l a r value of the services used i n a h o s p i t a l by MVA v i c t i m s , i t i s necessary to draw together the s p e c i f costs of the work units used to t r e a t the p a t i e n t s . For t h i s purpose, as d e t a i l e d i n the l i t e r a t u r e , the h o s p i t a l per diem rate, an average d a i l y value estimated from patient days, does not s p e c i f y d e t a i l e d cost f i g u r e s , e s p e c i a l l y at the work unit l e v e l . As suggested e a r l i e r i n t h i s report, the a l t e r n a t i v e i s to develop a step-down costing method that i d e n t i f i e s work un i t costs. - 39 -Step-down cost analysis begins with a breakdown of the t o t a l expenditure of the h o s p i t a l across departments/cost centres to i d e n t i f y i n d i v i d u a l departments/cost centres costs. This breakdown has been based on the H.S.2 return of the Vancouver General H o s p i t a l f o r 1975. Column (1) of Table D, headed " D i r e c t Expenses," records the reported expenditures by f u n c t i o n a l area from the H.S.2 forms and r e f l e c t s the operating expenses f o r 1975. The f u n c t i o n a l areas i d e n t i f i e d i n Table D are divided into d i r e c t p a tient care areas and i n d i r e c t p a tient services areas, and thus the t o t a l expenditures associated with, e.g., the nursing u n i t s , include both the d i r e c t expense a l l o c a t e d to those units and the i n d i r e c t expense required f o r heating, maintenance, and housekeeping of the nursing areas, as w e l l as meals, l i n e n , and employee benefits ( p r i m a r i l y pension con-t r i b u t i o n s and other p a y r o l l deductions associated with wages and s a l a r i e s generated i n the nursing u n i t s ) . Employee benefits are recorded separately i n lump sum form on the H.S.2 return form and are not charged back to the i n d i v i d u a l departments. In a step-down costing technique, the a l l o c a t i o n scheme takes the t o t a l expenditure associated with that service established f o r each i n d i r e c t or overhead area and then d i s t r i b u t e s i t to each f u n c t i o n a l area r e c e i v i n g overhead s e r v i c e s . For example, from Table D, the t o t a l plant overhead expense i s a l l o c a t e d according to the square footage of each department/cost centre (excluding unassigned areas such as XABUi. D r u n c t i o n a l Areas Step-Down Cost A l l o c a t i o n Plant Overhead Equipment Depreciation Employee Benefits Housekeeping Laundry/Linen Dietary Drugs Medical Supplies C.S.R. Medical Records Nursing Administration Nursing Units Emergency O.P.D. OR - PAR E.E.G. E.C.G. ICU Renal Audiology Psychology Obs./Cyn-Dellvery Nursery Rad tology Film Laboratory Phys. Med. & Rehab. Social Service Motor Service Education General Administration Pharmacy Service to Community Medical Staff Grandvlew Nursing Home Total 384922. 81848791. Direct Plnnt Equipment Employee House-Laundry/ Expense Overhead Depreciation Benefits Keeping Linen 2886199. 2945918.10 (59719.20) 1516349. 1516349.00 3887679. 67166.93 23351.77 4062530.40 3468471. 9704.63 211251.58 3698820.30 (9824.23) 1643384. 88082.95 33208.04 89375.67 110594.55 1964845.10 7318333. 213579.06 77030.53 301439.75 268164.05 19648.45 2169665. 3208240. 594452. 29164.59 15011.86 39406.54 36618.26 1105786. 68345.30 24109.95 73531.80 85512.50 1634939. 24745.71 108875.81 31070.04 19043366. 1216074.90 481137.53 1333728.70 1526870.60 1218203.90 1624970. 26218.67 23655.04 108469.56 32919.45 45191.44 606313. 70112.85 39121.80 40219.05 88031.79 68769.58 3959240. 115185.39 108267.31 263658.22 144623.65 589453.53 203029. 6481.02 7126.84 13812.60 8137.39 1964.85 557600. 6775.61 7278.48 30062.72 8507.27 2947.27 553084. 20621.43 14405.32 25891.70 869715. 15613.37 20015.81 58094.18 19603.72 72048. 1178.36 4468.78 1479.53 63612. 589.18 1625.01 739.76 607732. 139047.33 48523.17 87344.40 174584.04 702938. 10605.31 13315.73 2530973. 86315.40 136926.31 138938.53 108375.26 4912.11 260001. 5904948. 132271.72 152999.61 319314.88 166076.77 1325828. 61569.69 26232.84 87344.40 77305.23 3929.69 593455. 11489.08 8946.46 39406.54 14425.38 578583. 38187.79 4680683. 332888.74 159974.81 240908.05 417966.04 6817411. 179995.59 77333.80 342877.56 225997.57 474643. 21799.79 22138.70 30875.23 27371.23 Dietary Drugs Hed.Supplies t CSR 8197504.00 5266896.30 2169665.00 230380.83 49379.21 97235.81 419072.00 48666.11 196549.76 63.28 1125825.34 2945918.10 1516500.50 (.0099X) error 4062936.60 (.0099X) error 3698814.40 1964845.15 50824.52 t 5317720.80 (2879784) To employees' meals 3925385.70 1374277.50 164866.19 2041985.60 81255.48 236308.21 1177.62 3925778.10 (.0099X) error Hedleal fesax&a. 1357585.60 773823.73 54303.42 271517.10 1357585.60 Nursing Admin. Sub-Total General Admin. 1799630.50 1320748.80 33785511.00 4010112.70 37795623.00 82603.04 2212575.00 262576.80 2475151.80 57768.14 1359803.50 161202.20 1521005.70 280742.35 7922228.10 940623.20 8862851.30 240551.70 28251.90 268803.60 613171.36 73122.70 686294.10 16556.60 760480.64 90572.40 851053.00 32933.24 1448833.20 172004.40 1620837.60 79174.67 9140.33 88315.00 66565.95 7893.92 74460.00 1057230.90 125421.82 1182652.70 726859.04 86417.68 813276.70 3006440.60 *260001.0O 356472.92 3362913.50 (260001.00) 6675611.00 . 791885.08 7467466.10 1583450.70 187792.26 1771243.00 667722.50 78939.23 746661.70 616770.80 to Gen.Admin. 6049634.30 717931.49 6767565.80 8278.30 7651893.90 to Gen.Admin. 1707756.20 50824.52 202749.17 5816.57 1910505.40 56641.10 40727.57 to Gen.Admin. fo84922.00) (384922.00) 1799630.30 '(78323320.00) 8308925.80 *78323320.00 excluded from Gen. Admin, allocation tt - 41 -co r r i d o r s and s t a i r s , or areas assigned to plan maintenance and operation) as reported i n A l l o c a t i o n Formula A. (See page 42). A d i f f i c u l t y a r i s e s because i n d i r e c t patient s e r v i c e areas supply services to each other. Plant Overhead, for example, includes heating and maintenance f o r the Dietary and Laundry areas, while these departments supply services to Plant Overhead i n the form of employee meals and employee uniforms r e s p e c t i v e l y . The usual procedure i n step-down technique i s to arrange areas i n a h i e r a r c h i c a l order and then a l l o c a t e areas supplying services to others but r e c e i v i n g none i n return, to services lower i n the order and then c l o s i n g down these areas, having a l l o c a t e d a l l the departments' costs across other d i r e c t patient care departments or cost centres. Once done, these i n d i r e c t patient services can be considered f u l l y a l l o c a t e d and the next step i s to develop work unit costs for each d i r e c t p a t i e n t s e r v i c e . The problem with t h i s approach i s that the matrix i s not h i e r a r c h i c a l . For example, Plant Overhead accounts for a portion of employee bene f i t s associated with the s a l a r i e s of plant and maintenance workers. But employee ben e f i t s include p h y s i c a l f a c i l i t i e s such as lounges and locker rooms and lunchrooms which generate plant overhead expense. These same areas also generate housekeeping expenses while the s a l a r i e s of housekeeping personnel generate employee b e n e f i t s . To complete t h i s - 42 -ALLOCATION FORMULA "A" SQUARE FOOTAGE OF FLOOR AREA % Nursing Units 466,261 41.28 Nursing Administration 9,509 .84 Renal 6,025 .53 ICU 7,962 .70 EEG 2,450 .22 ECG 2,560 .23 Psychiatry 227 .02 Audiology 453 .04 Nursing 4,075 .36 Obs/Gyn 53,312 4.72 OR-PAR 44,046 3.91 Emergency 10,009 .89 Outpatient Department 26,898 2.38 Physical Medicine 23,562 2.09 Laboratories 50,710 4.49 Radiology 33,054 2.93 Pharmacy 8,367 .74 Social Services 4,361 .39 Dietary 81,901 7.25 Laundry 33,809 2.99 C . S • R • 11,227 .99 Employee Benefits 25,745 2.28 Medical Records 26,233 2.32 Education 127,634 11.30 General Administration 68,953 6.11 1,129,443 100 Central Stores 34,985 Building Maintenance Housekeeping 145,484 not a l l Stair Corridors, Elevators 263,738 ocated - 43 -interconnection, the laundry and l i n e n area supplies service to both Housekeeping and Dietary areas, while Housekeeping services are supplied to both Laundry/Linen and Dietary. A l l areas generate plant overhead expense and a l l s a l a r i e s generate employee b e n e f i t s , completing the i n t e r a c t i o n between these s i x departments. Only the expenditures associated with equipment depreciation are i s o l a t e d from a l l other i n d i r e c t p atient areas. Service areas with large amounts of equipment may generate s i g n i f i c a n t depreciation expense, but the depreciation items generate no expenditure i n return. This problem of simultaneous flow back and f o r t h between these services i s resolved by s e t t i n g up a block of simultanous equations that allows, i n e f f e c t , the c l o s i n g down of a l l i n t e r l o c k i n g overhead departments at one time. Each overhead area i s a l l o c a t e d , following 2 the same process ou t l i n e d i n Evans and Robinson. A l l o c a t i o n of Overhead Departments Plant Overhead expenditures (plus a share of employee benefits) are a l l o c a t e d according to A l l o c a t i o n Formula "A," page 42. From t h i s schedule, the apportionment of f l o o r area accounted f o r by f u n c t i o n a l departments excludes non-assigned space ( s t a i r s and elevators) plus c e n t r a l stores and b u i l d i n g maintenance and housekeeping. This has the e f f e c t of spreading the cost of these non-assigned areas over the included f u n c t i o n a l departments i n proportion to the square footage Evans R.G., and Robinson G.C., op.cit., 32. - 44 -a l l o c a t e d each department. The proportions c a l c u l a t e d i n A l l o c a t i o n "A" are then m u l t i p l i e d by $2,945,918.00 to a r r i v e at the en t r i e s i n Column (2) of Table D. Small deviations i n percentage between the t o t a l of these e n t r i e s ( l a s t l i n e ) and the i n d i v i d u a l f i g u r e s to be a l l o c a t e d are due to round-off e r r o r . The a l l o c a t i o n formula f o r major depreciation, A l l o c a t i o n Formula "B", page 45, i s more complex to develop. The d o l l a r value from the H.S.2 return f o r Vancouver General H o s p i t a l i s a percentage of the t o t a l year's expenditure which i s then a l l o c a t e d to a l l b u i l d i n g s and equip-ment on the campus of the h o s p i t a l . The H.S.2 return shows b u i l d i n g depreciation at $1,030,387.00 and equipment depreciation at $485,962.00. The h o s p i t a l was not i n a p o s i t i o n to furth e r i d e n t i f y e i t h e r which bu i l d i n g s the depreciation should be a l l o c a t e d to or which pieces of equipment should carry the equipment depreciation costs. The b u i l d i n g depreciation f i g u r e was a l l o c a t e d to a l l departments/cost centres on a square footage b a s i s , as i d e n t i f i e d i n A l l o c a t i o n Formula "A." The equipment depreciation needed more d e f i n i t i o n before i t could be a l l o c a t e d across the departments. I t was decided to average the equipment expense of each department/cost centre over the l a s t three years, 1973, 1974 and 1975,to determine each department's share of c a p i t a l expenditures by the h o s p i t a l . This percentage f i g u r e f o r each department i s seen i n Column (2) of A l l o c a t i o n Formula "B." The - 45 -ALLOCATION FORMULA " B " B u i l d i n g D e p r e c i a t i o n A l l o c a t i o n $ 1 , 0 3 0 , 3 8 7 . Equipment D e p r e c i a t i o n Percentage of U.S.2 Value  C a p i t a l Cost $ 4 8 5 , 9 6 2 . Nurs ing U n i t s 4 3 2 , 5 5 6 . 4 6 10% 48,596 . 2 0 Renal 5 , 4 6 1 . 0 5 3% 14 ,528 . 8 0 ICU 7 , 2 1 2 . 7 1 •1.5% 7,264 .40 EEG 2 , 2 6 6 . 8 5 1% 4 ,859 . 6 2 ECU 2 , 3 6 9 . 8 9 1% 4,859 .62 OR/PAR 4 0 , 1 8 5 . 0 9 14% 68,034 .68 Obs/Gyn 4 8 , 5 3 1 . 2 3 Emergency 9 , 0 6 7 . 4 1 i 3% 14,528 .80 P h y s i c a l M e d i c i n e 2 1 , 4 3 2 . 0 5 1% 4,859 62 OPD 2 4 , 5 2 3 . 2 1 3% 14 ,528 80 Lab 4 6 , 1 6 1 . 3 4 1 22% 106 ,911 64 X - r a y 3 0 , 0 8 7 . 3 0 22% 1 0 6 , 9 1 1 . 64 Pharmacy 7 , 6 2 4 . 8 6 1 3% 1 4 , 5 2 8 . 80 Soc. S e r v i c e s 4 , 0 1 8 . 5 1 ' 1% 4 , 8 5 9 . 62 D i e t a r y 7 4 , 6 0 0 . 0 2 1 .5% 2 , 4 2 9 . 81 Housekeeping 7 , 2 1 2 . 0 9 . 5 % 2 , 4 2 9 . 81 Laundry 3 0 , 8 0 8 . 5 7 . 5 % 2 , 4 2 9 . 81 C . S . R . 1 0 , 2 0 0 . 8 3 1% 4 , 8 5 9 . 62 Emp. B e n e f i t s 2 3 , 3 8 9 . 7 8 Med ica l Record 2 4 , 1 1 1 . 0 6 Educat ion 1 1 6 , 2 2 7 . 6 5 9% 4 3 , 7 3 6 . 58 Gen. Admin. 6 2 , 7 5 0 . 5 6 3% 1 4 , 5 2 8 . 80 T o t a l Percentage 4 8 1 , 1 5 2 . 6 6 3 1 . 7 3 1 9 , 9 8 9 . 8 5 1 .32 1 4 , 4 7 7 . 1 1 . 9 5 7 , 1 2 6 . 4 7 .47 7 , 2 2 9 . 5 1 . 4 8 1 0 8 , 2 1 9 . 7 7 7.14 4 8 , 5 3 1 . 2 3 3 . 2 0 2 3 , 5 9 6 . 2 1 1 .56 2 6 , 2 9 1 . 6 7 1 . 7 3 3 9 , 0 5 2 . 0 1 2 . 5 8 1 5 3 , 0 7 2 . 9 8 10.09 1 3 6 , 9 9 8 . 9 4 9 . 0 3 2 2 , 1 5 3 . 6 6 1 . 4 6 8 , 8 7 8 . 1 3 .59 7 7 , 0 2 9 . 8 3 5 . 0 8 9 , 6 4 2 . 5 2 .64 3 3 , 2 3 8 . 3 8 2 .19 1 5 , 0 6 0 . 4 5 .99 2 3 , 3 8 9 . 7 8 1.54 2 4 , 1 1 1 . 0 6 1 .59 1 5 9 , 9 6 4 . 2 3 1 0 . 5 5 7 7 , 2 7 9 . 3 6 5 . 1 - 46 -ALLOCATION FORMULA "C" EMPLOYEE BENEFITS & SALARIES % Nursing Administration 1,632,422 2. 68 Nursing Units (plus students) 19,998,205 32. 83 Psychology 24,236 04 Audiology 66,441 11 C • S . R • 593,554 97 Nursery Obs. 1,310,670 2. 15 Renal 869,449 1. 43 OR/PAR 3,952,597 6. 49 Emergency 1,624,565 2. 67 Outpatients 600,422 . 99 Lab 4,789,147 7. 86 ECG 451,222 74 EEG 204,122 34 Pharmacy 462,896 76 Radiology 2,085,958 3. 42 Physical Medicine 1,309,039 2. 15 Social Services 592,511 97 Education 3,612,207 5. 93 Dietary 4,517,857 7. 42 Laundry 1,343,287 2. 20 Housekeeping 3,165,414 5. 20 Motor Services 574,118 , 94 Plant Operation and Maint. 897,832 1. 47 Medical Records and Library 1,104,128 1. 81 General Administration 5,139,652 8. 44 60,921,951 100. 01 - 47 -i n d i v i d u a l percentages were then applied to the equipment depreciation cost of $485,962.00, to a l l o c a t e the t o t a l cost among the member departments/cost centres. The b u i l d i n g depreciation and equipment depr e c i a t i o n costs f o r each department were then added together to determine the departmental share of t o t a l d epreciation of the h o s p i t a l . This i s out l i n e d i n Column (4) of A l l o c a t i o n Formula "B." Employee b e n e f i t s included $3,887,679.00 of d i r e c t expense f o r Vancouver General H o s p i t a l on behalf of i t s s t a f f , as w e l l as plant overhead and housekeeping costs f o r areas of the h o s p i t a l designated f o r employee use and the depreciation of these areas. A t o t a l of $4,062,530.40 has been a l l o c a t e d according to A l l o c a t i o n Formula "C," which i s the share of each f u n c t i o n a l department i n the t o t a l h o s p i t a l wage and sal a r y payments. The f i n a l d o l l a r f i g u r e f o r employee ben e f i t s i s found i n Column (4), top l i n e i n Table D. Housekeeping costs were ca l c u l a t e d f rom the t o t a l of d i r e c t expense of $3,468,471.00 from the H.S.2 return plus depreciation, employee ben e f i t s and a p o r t i o n of laundry and l i n e n expenses f o r maintenance of uniforms, t o t a l l i n g $3,608,820.30. This t o t a l i s a l l o c a t e d i n Column (5) of Table D, according to Schedule "A," h o s p i t a l f l o o r area. Laundry and l i n e n services cost $1,964,845.10, comprised of a d i r e c t expense of $1,643,584.00 plus employee b e n e f i t s , plant overhead, - 48 -housekeeping and depreciation. The appropriate d i s t r i b u t i o n f o r t h i s t o t a l would i d e a l l y be by pounds of laundry processed f o r each f u n c t i o n a l department; since t h i s information was not a v a i l a b l e , an estimated a l l o c a t i o n scheme was constructed i n c o n s u l t a t i o n with the Vancouver General Hospital's laundry manager. This estimated d i s t r i b u -t i o n i s given i n A l l o c a t i o n Formula "D" and the r e s u l t i n g a l l o c a t i o n of laundry expense i s recorded i n Column (6) of Table D. Dietary department t o t a l expense of $8,197,504.00 i s comprised of $7,318,333.00 of d i r e c t expense plus plant overhead, housekeeping, employee b e n e f i t s , and laundry s e r v i c e s . Meals were supplied to three areas, nursing u n i t s , s t a f f c a f e t e r i a s and i n 1975, two other i n s t i t u -tions outside the H o s p i t a l , as i s seen i n the H.S.I return. From discussions with the accounting department of Vancouver General H o s p i t a l , the cost of s t a f f meals was determined by d i v i d i n g the revenue from the s t a f f c a f e t e r i a by the number of s t a f f meals i n the year, to come up with an average meal cost of $1.52. This was approved by both the accounting department and the Province. Therefore, the r e a l costs of the dietary department would be that which i s not recovered (the meals to the nursing u n i t s , 64.25% of the meals, and those to two other i n s t i t u t i o n s , .61% of t o t a l meals). See Column (7), Table D for the d o l l a r breakdown of the d i e t a r y department. Table D shows that costs f o r employee meals were excluded from the a l l o c a t i o n of dietary expenses. These costs w i l l also be excluded from the c a l c u l a t i o n s of per diem - 49 -ALLOCATION FORMULA "D" ESTIMATED DISTRIBUTION OF LAUNDRY & LINEN EXPENSE BY FUNCTIONAL AREA % Nursing Units 62 OR/PAR 30 Emergency 2.3 Outpatient 3.5 Dietary .1 Rehab and P h y s i c a l Medicine .2 Radiology .25 EEG •10 ECG .15 Housekeeping .5 100.00 - 50 -costs of a l l h o s p i t a l services used l a t e r i n t h i s report. Drugs, Medical/Surgical Supplies, CSR V.G.H. had, i n 1975, a cost centre accounting system f or t h e i r i n t e r n a l operations. This system aided g r e a t l y i n the a l l o c a t i o n of the next two items, drugs and CSR. Column (8), Table D, reported only the a l l o c a t i o n of the costs of the actu a l drugs purchased by V.G.H. For example, ward stock to the nursing units cost $230,380.83 while drugs used i n the Pharmacy cost $1,125,825.34. Therefore, step-down of drug expense follows the d i s t r i b u t i o n i n d i c a t e d i n the cost centre printouts from V.G.H. as reported i n A l l o c a t i o n Formula "E". P r e s c r i p t i o n costs f o r patient services provided by the Pharmacy w i l l be reviewed l a t e r i n t h i s chapter when a u n i t cost f o r patient p r e s c r i p t i o n s w i l l be developed. Overhead costs r e l a t e d to drug d i s t r i b u t i o n and purchase w i l l be accounted f o r i n the unit value f o r Pharmacy services,.as they are responsible f o r the management and d i s t r i b u t i o n of drug supplies. Rows (9) and (10) of Table D, medical/surgical supplies and c e n t r a l supply room, expenditures were consolidated and a l l o c a t e d i n Column (9) of Table D. The costs included the d i r e c t expenses f o r C.S.R. operation and f o r the medica l / s u r g i c a l supplies, and the i n d i r e c t costs associated with operating the C.S.R. (I t i s possible to argue that ALLOCATION FORMULA "E" DRUG EXPENSE FROM THE COST CENTRE PRINTOUT Nursing Units T o t a l l e d Renal ICU Emergency Outpatient Department OR/PAR CSR P h y s i c a l Medicine Pharmacy $ 2 3 0 , 3 8 0 . 8 3 1 9 6 , 5 4 9 . 7 6 4 8 , 6 6 6 . 1 1 4 9 , 3 7 9 . 2 1 9 7 , 2 3 5 . 8 1 4 1 9 , 0 7 2 . 0 0 2 , 4 9 2 . 7 2 63 .28 1 , 1 2 5 , 8 2 5 . 3 4 - 52 -since the C.S.R. also handled a share of drugs, part of i t s expenses should be a l l o c a t e d p r o p o r t i o n a l l y to the flow of drug expense. But t h i s i s such a small part of the t o t a l that the extra gained i n accuracy would be t r i v i a l compared to the workload involved.) As i n the case of drug expenses, Column (9) i s a l l o c a t e d , using A l l o c a t i o n Formula "F", which i s based on the pri n t o u t from the cost centre system. These costs included the materials i d e n t i f i e d f o r the C.S.R. plus the d i r e c t expense from the H.S.2 return f o r the C.S.R., the i n d i r e c t overhead expenses and, of course, the medical/surgical supplies. The medical records department expense t o t a l l e d $1,357,585.60 based on d i r e c t expense of $1,105,786.00 and the i n d i r e c t overhead costs. Through co n s u l t a t i o n with the medical records department d i r e c t o r , an estimated d i s t r i b u t i o n of time spent i n th i s department to s a t i s f y the needs of other f u n c t i o n a l areas was determined. This was given i n A l l o c a t i o n Formula "G", and the r e s u l t i n g a l l o c a t i o n of medical record expenses i s recorded i n Column (10) of Table D. Nursing Administration costs are a l l o c a t e d across the various f u n c t i o n a l departments by the number of supervisors, head nurses and t h e i r a s s i s t a n t s . A t o t a l of $1,799,630.50 was included i n t h i s cost. Assuming that administration demands are comparable f o r each department, the v a r i a t i o n i n demand should mirror the number of people required to handle that aspect of nursing. Therefore, t h i s a l l o c a t i o n as shown i n Column (11), - 53 -ALLOCATION FORMULA "F" SURGICAL AND OTHER MEDICAL SUPPLIES Nursing Units $1,018,321.20 35.01 OR/PAR 1,513,314.80 52.02 ICU 60,125.29 2.07 Renal 175,076.70 6.02 Emergency 122,187.26 4.20 Physical Medicine 827.84 0.03 OPD 15,438.81 0.53 Pharmacy 3,766.24 0.13 2,909,057.90 100.01 CSR 286,552.57 - not allocated - 54 -ALLOCATION FORMULA "G" MEDICAL RECORD DEPARTMENT SERVICES % Inpatient Units 57 Outpatient C l i n i c 17 Emergency Department 4 Su r g i c a l Outpatients 3 Education 8 Research 8 Medical S t a f f A c t i v i t i e s 3 - 55 -Table D, i s based on the A l l o c a t i o n Formula "H", page 56. Column (12), Table D, shows a subtotal f o r those f u n c t i o n a l areas that remain unallocated. Several of these can be argued as i n d i r e c t patient s e r v i c e s , s p e c i f i c a l l y , general administration, motor s e r v i c e and medical s t a f f administration costs. The l a s t two items were added to general administration, which i n turn was a l l o c a t e d across the remaining f u n c t i o n a l areas. The a l l o c a t i o n formula used to d i s t r i b u t e t h i s new general administration cost was based on the r a t i o of each department's budget to the t o t a l a l l o c a t e d budget. To be accurate, the new general a d m i n i s t r a t i o n cost should be subtracted from the t o t a l a l l o c a t e d budget before these percentages are c a l c u l a t e d . For example: Nursing Unit Budget  (Total A l l o c a t e d Budget - New General X N e w G e n e r a l = The share of new Administration Budget) Admin Budget general budget allocated to Nursing Units. 33,785,511.00 X 8,308,925.80 (78,323,320 - 8,308,925.80) = 48.26% or $4,010,112.75 - 56 -ALLOCATION FORMULA "H" NURSING ADMINISTRATION Number of Administration Functional Area Staff  Nursing Units OR PAR Emergency 0. P.D. Renal 1. C.U. General Administration 160 73.39% 34 15.60% 10 4.59% 7 3.21% 4 1.83% 2 .92% 1 .46% 218 100 - 57 -As i s noted i n Table D, the f u n c t i o n a l areas of Grandview Nursing Home and radiology f i l m were not included i n t h i s l a s t a l l o c a t i o n because: (1) Grandview Nursing Home was on the H.S.2 return form f o r Vancouver General H o s p i t a l purely as an accounting operation and thus had no r e a l involvement i n Vancouver General Hospital's day to day operations; and (2) Radiology f i l m i s , l i k e drugs, a primary raw mate r i a l expense and did not draw on any services of the H o s p i t a l ; those services necessary to incorporate radiology f i l m i n t o the day to day operations, i . e . , administration, maintenance, storage, etc., would be incurred by the radiology department as seen i n Line (24), Table D. The t o t a l of Column (14) i s less than the t o t a l of Column (1), due to the exclusion of the values of employees' meals, radiology f i l m and Grandview Nursing Home. I f these values were added to the t o t a l of Column (14), t h i s t o t a l would be $81,848,027.00, only .00094% e r r o r . - 58 -Unit Values The c a l c u l a t i o n of unit values f o r the remaining f u n c t i o n a l departments ranges from exact d i v i s i o n s of the t o t a l d o l l a r value by u t i l i z a t i o n , to estimates based on consultations with s t a f f of those f u n c t i o n a l departments. The following i s a l i s t of the f u n c t i o n a l areas remaining, as seen i n Column (14), TableD . 1. Nursing Units 2. Emergency 3. Operating Room 4. EEG 5. ECG 6. ICU 7. Renal 8. Audiology 9. Psychology 10. Obstetrics/Gynecology/Delivery 11. Nursery 12. Radiology 13. Film 14. Laboratory 15. P h y s i c a l Medicine 16. S o c i a l Service 17. Education 18. Pharmacy Not a l l f u n c t i o n a l areas p e r t a i n to MVA pat i e n t s , but those that do w i l l be determined from the analysis of the medical records of the MVA vi c t i m s . The following i s a breakdown of the method used to determine the u n i t values of these f u n c t i o n a l areas. -59-Nursing Units Since s a l a r y and overhead costs are stepped-down across a l l nursing u n i t s , the best approach to defining a u n i t cost would be to i d e n t i f y the t o t a l proportion of resources used by each d i v i s i o n of nursing care ( i . e . , m e d i c a l / s u r g i c a l , o b s t e t r i c s , p a e d i a t r i c s , convalescent and extended care, i n c l u d i n g chronic care). From Table D, the t o t a l a l l o c a t e d budget f o r nursing units i s $37,795,623.00. By f a r the la r g e s t component of t h i s cost i s for nursing s a l a r i e s ; thus the d i s t r i b u t i o n of f u l l - t i m e equivalent nursing s t a f f to each of these l e v e l s of care was used to a l l o c a t e the resources to a l l nursing u n i t s . In 1975 nursing s t a f f were on a geographic basis to p a v i l i o n s , rather than by l e v e l s of care. But the H.S.2 form i d e n t i f i e s nursing s a l a r i e s a l l o c a t e d to short term u n i t s (adult and c h i l d ) and to long term u n i t s of $17,250,584 and $1,785,950 r e s p e c t i v e l y . Without a patient c l a s s i f i -c a tion system, a basic assumption was made that nursing FTE's were a l l o c a t e d i n a s i m i l a r manner to the component l e v e l s of care that were offe r e d i n the short term u n i t . This assumption was made f o r the two l e v e l s of care that were offered i n the long term u n i t . A second assumption was made that labour, d e l i v e r y and nursery costs must be added to the o b s t e t r i c s t o t a l to ensure that t h e i r operating costs were included i n t h i s study. Short Term Bed D i s t r i b u t i o n (set up December 31, 1975) Medical/Surgical beds - 1,218 83.0% P a e d i a t r i c beds - 167 11.3% Ob s t e t r i c beds - 83 5.6% - 60 -Using these percentages, the nursing s a l a r i e s f o r short term units ($17,250,584) can be a l l o c a t e d as: Medical/Surgical -$14,334,662 P a e d i a t r i c s - 1,947,834 Obstetrics - 968,084 Labour and Delivery - 607,737 Nursing - 702,938 Long Term Bed D i s t r i b u t i o n R e h a b i l i t a t i o n beds - 47 16.2% Extended (Chronic) beds - 242 83.8% Using these f i g u r e s , the nursing s a l a r i e s a l l o c a t e d to long term units ($1,785,950) can be a l l o c a t e d as: R e h a b i l i t a t i o n -$ 290,449 Extended - 1,495,501 Thus, to a l l o c a t e the nursing unit costs of $37,795,623, Table D, the d i s t r i b u t i o n of nursing s a l a r i e s as c a l c u l a t e d above were used. Level of Care S a l a r i e s Percentage A l l o c a t e d Dollars Medical/Surgical 14,334,662 70.45% $26,627,016 Pa e d i a t r i c s 1,947,834 9.56% 3,617,041 Obste t r i c s 2,278,759 11.20% 4,233,110 R e h a b i l i t a t i o n 290,449 1.43% 540,477 Extended (Chronic) 1,495,501 7.35% 2,777,978 To get a d a i l y cost f o r each of these l e v e l s of care, t h e i r a c t u a l separation days f o r 1975 were used. This accounted for those bed days not used when the occupancy of any of these units was below 100%. - 61 -Patient separation days by l e v e l of care were a v a i l a b l e from the H.S.I form. Intensive care unit patient separation days were separated from me d i c a l / s u r g i c a l patient separation days and w i l l be reviewed l a t e r i n t h i s report. Level of Care Patient Separation Days 408,037 A l l o c a t e d D o l l a r s D a i l y Cost Medical/Surgical $26,627,016 $ 65. 26 P a e d i a t r i c s 55,486* 3,617,041 65. 19 Obs t e t r i c s 26,977 4,233,110 156. 92 R e h a b i l i t a t i o n 13,370 540,477 40. 42 Extended (Chronic) 98,795 2,777,978 28. .12 Source: A l l data de f i n i n g patient separation days and beds set up f o r each l e v e l of care come from 1975 Annual Return of H o s p i t a l - Form H.S.I for Vancouver General H o s p i t a l . Occupancy of P a e d i a t r i c beds (167 beds set-up) appears to be high, but data r e f l e c t s f igures used by H o s p i t a l i n 1975 H.S.I form. - 62 -Emergency The determination of a unit value f o r emergency services needed a weight f a c t o r that r e f l e c t e d the s e v e r i t y of the patient's i n j u r y . The t r a d i t i o n a l c l a s s i f i c a t i o n system used i n t h i s department i s t r i - l e v e l , i . e . , emergent, urgent and non-urgent. Because of the department's r o l e i n the health d e l i v e r y network, a s i g n i f i c a n t amount of s t a f f s k i l l s and equipment resources i s oriented to the emergent patient, but used in f r e q u e n t l y compared to the use of the department by non-urgent patients'; In 1975, V.G.H. emergency department s t a f f was a l l o c a t e d i n the following manner:^ 0 50% of s t a f f to emergent s t a t i o n s ; • 27.9% of s t a f f to urgent s t a t i o n s ; and • 22.1% to non-urgent s t a t i o n s When the t o t a l 67,690 patient v i s i t s were c l a s s i f i e d , only 4,330 were emergent and 7,785 were urgent; the remaining patients were c l a s s i f i e d as non-urgent. Following the same l o g i c used f o r other V.G.H. departments i n developing unit costs, emergent patients should have consumed 50% of department resources ($2,475,151.80 from Table D) as 50% of s t a f f were assigned to emergent s t a t i o n s . Therefore a u n i t cost f o r the 4,330 patients would be $285.80. Urgent patients t o t a l l i n g 7,785 would use 27.9% of Conversations with Head Nurse - Emergency Department - 63 -resources or have a u n i t cost of $88.70. Non-urgent patients would have a u n i t cost of $9.85. While t h i s process produced a u n i t d o l l a r value, i t i s questionable whether i t r e f l e c t e d the actual use of the department by the patients i n each of these three l e v e l s . For example, the s t a f f assigned to emergent cannot p o s s i b l y spend a l l t h e i r time on emergent pa t i e n t s . Some of t h e i r time must be spent on general routine or t r e a t i n g non-urgent patients using resources of the emergent workstations. Without a "time and motion" study, i t was impossible to further r e f i n e t h i s c l a s s i f i c a t i o n system. F i n a l l y , the patient medical record does not c l a s s i f y the patient as emergent, urgent or non-urgent at the time of his/her a r r i v a l to the emergency depart-ment, therefore the weight f a c t o r was of no value to t h i s study. Other methods of weighting the demands of patients on the department's resources, such as time spent i n department, were equally of l i t t l e value. Given that a l l a n c i l l a r y and diagnostic t e s t s and s e r v i c e s , s p e c i f i c a l l y radiology, laboratory and OR time, used by the patient while i n the emer-gency department w i l l be i d e n t i f i e d and an appropriate cost assigned to the patient through the diagnostic department/cost centre, i t was f e l t that an average cost, could be assigned to a l l patients using the emergency department. In 1975 the patient load i n t h i s department was 67,690 and the a l l o c a t e d budget from Table D was $2,475,151.80, r e s u l t i n g i n an average value f o r an emergency v i s i t of $38.60. - 64 -This approach implies that the same basic resources were a l l o c a t e d f o r high and low s e v e r i t y patients using t h i s department. Because the emergency needs of MVA victims could not be r e t r o s p e c t i v e l y categorized, a l l was not l o s t i n using a department average. The non-urgent patient cost increased from $9.85 to $38.60 f o r a d i f f e r e n c e of $28.75. The emergent patients u n i t cost decreased from $285.80 to $38.60 for a di f f e r e n c e of $247.20. Therefore, the regression equation i n Chapter TV of t h i s t h e s i s , which analyses the r e l a t i o n s h i p of cost with severity i s r e s t r i c t e d , e r r i n g on the conservative side. I f a s u i t a b l e measure was made a v a i l a b l e to i d e n t i f y emergent, urgent, and non-urgent patients and an appropriate . weighted u n i t cost, then the regression equation would be made more s e n s i t i v e than the r e s u l t s i n Chapter IV i n d i c a t e . - 65 -Operating Room Operating room unit costs were derived by using time as the weighting f a c t o r . P h ysician costs on a fee f o r service basis were not included i n t h i s study. Therefore, only those costs of operating the s u r g i c a l s u i t e s at Vancouver General H o s p i t a l were considered. The use of operating room f a c i l i t i e s was averaged f o r a l l procedures performed at the H o s p i t a l i n 1975. I t i s possible that the short term operating room procedures may take more services than the cost a l l o c a t e d to them by the time of the operation. On the other hand, longer term cases could be more severe than short term operations, and thus on the average, use more services than a l l o c a t e d by the time f a c t o r . The major cost component i s s t a f f cost. Therefore, time i s the only v a l i d comparison across cases. The use of sk i n - t o - s k i n time assumes that preparation and recovery time i s proportional to a c t u a l operating room time. In Vancouver General H o s p i t a l there are over 30 operating rooms; the assumption was made that a l l operating rooms are the same. The one procedural drawback was that Vancouver General H o s p i t a l did not keep a running t o t a l of time that i t s operating rooms were i n use. To overcome t h i s , records were kept during 4 one-week periods during 1975; the times were m u l t i p l i e d by 13 to get an i n d i c a t i o n of the year's t o t a l time. The 4 one-week periods were chosen to include seasonal e f f e c t s , vacations and statutory holidays so that the - 66 -estimated u t i l i z a t i o n f igures f o r 1975 would be representative of the ac t u a l f i g u r e s . Fluctuations across the four t o t a l s i n the following chart were expected. These f l u c t u a t i o n s r e i n f o r c e the need to choose representative weeks of the year that r e f l e c t f a c t o rs which influence u t i l i z a t i o n ( i . e . , s tatutory holidays and t r a d i t i o n a l s t a f f vacation time periods). The week with the statutory holiday (May 18-24) had the lowest use, while the weeks representing f a l l and winter had the highest use, suggesting that during t h i s time of year most s t a f f and physicians were a v a i l a b l e f o r operations. The week representing summer had a low use, r e f l e c t -ing the concern that both s t a f f and patients would be on holidays. Operating Room Feb.9-15 May 18-24 Aug.10-16 Nov.16-22 W.C.T.U. 3,229 1,937 1,434 3,757 W.P.O.R. 818 766 608 694 H.P.U.B.R. 2,038 1,540 1,675 2,044 C.P.O.R. 12,919 9,492 10,940 12,186 CP.A.R. 873 700 657 721 H.P.O.R. 11,231 8,850 9,558 12,624 D.O.R. 2,788 1,672 1,903 2,204 To t a l 33,896 24,957 26,775 34,230 Grand T o t a l 119,858 min. - 67 -To account f o r a f u l l year, the four-week sample of 119,858 minutes was m u l t i p l i e d by 13 to get a p r o j e c t i o n of 1,558,154 minutes of operating room time f o r 1975. Since the OR-PAR costs from Table B were $8,862,851.30, the cost of OR-PAR time per minute of s k i n - t o -s k i n time i n the operating room was $5.69. This value was m u l t i p l i e d across the s k i n - t o - s k i n time noted on the MVA patient's operating room form i n his/her medical record. Electroencephalography The EEG department o f f e r s s i x electroencephalographic procedures. This presented a d i f f i c u l t y because the i n t e r n a l operation of t h i s department d i d not o f f e r a predetermined ranking of these s i x d i f f e r e n t procedures. To counteract t h i s , the fee schedule of the p r o v i n c i a l government was used as a ranking mechanism of the s i x ways of d e l i v e r i n g an e l e c t r o -encephalogram. The reader may object to t h i s method, as these d o l l a r f i g u r e s r e f l e c t the time needed by the physi c i a n to i n t e r p r e t the test. But since 25% —30% of the department's budget i s for medical s t a f f and following advice from the chief physician of the department, the fee schedule was used. This allowed one approach f o r a weighting system to be developed that ranked the component t e s t s of t h i s department. The foll o w i n g i s the process f o r a r r i v i n g at the un i t value f o r the s i x pro-cedures of the EEG department. - 68 -Rank Based on Share of Department's Share of Fee Schedule Resources T n r n i . Procedure r> ^  , resources Total Average L e a u r e Rate Number D o n ^ r - o v * . . ,T . J= Unit Cost Resourcesllars %Total Budget 2,864 2 2 $129,796.48 226.60 199.84 94.07 0.16 0.14 252,863.54 430.09 376.33 14 1,484.00 1.08 2,903.08 8 79 2,969 1,016.96 5,253.50 .74 3.81 1,989.15 10,241.42 Regular EEG 45.32 ?m,/?o.mi y4.07 252,863.54 88.29 Electrocortiography 113.30 0.16 430.09 215.05 Speech Recordings 99.92 0.14 376.33 188.17 Sphenoid electrode Recordings 106.00 j.t i ,4o4.uu 1.08 2,903.08 207.36 Injections of Corti-cal stimulants 127.12 8 1,016.96 .74 1,989.15 248.64 ICL, VEL, EMOG 66.50 7Q * * °- 129.64 With sufficient manpower and time, the average drain on the EEG department for each of the above procedures could be determined and ranked proportionately. This was not feasible for this study. There seems to be 2:1 ratio between average unit cost and fee schedule cost (the same concerns are applicable for radiology - seen later in this section of the report). It is unlikely that the government fee schedule was developed so that physician costs would roughly equal a l l other departmental costs. Since 96% of a l l tests of this service were regular EEG and since those MVA included in this analysis received only regular EEG, another methodology for determining unit cost for this service would be to divide the cost ($268,803.60) by total tests (2,969) to get an average cost per test of $90.54. Using a weight factor did not change the unit cost significantly for this department in context of the demand placed on this department by MVA victims. Intensive Care Unit The ICU department offered a rather simple process for determining i t s unit cost. The medical records department was able to determine that the total patient load in the ICU in 1975 was 2,949 patient days. Therefore, with the total yearly cost of $851,053 from Table D, the unit cost was $288.60 per day. - 69 -Electrocardiography The ECG department provided p r i m a r i l y regular ECG services i n 1975; the only other service offered was a telephone service f o r t e s t i n g pacemakers. The l a t t e r s e rvice was ignored because i t s drain on the department was minimal (numbers t o t a l only about 50 c a l l s per month). Therefore, with the t o t a l cost of $686,294.10 and a t o t a l number of tests completed of 24,084, a unit cost per ECG would be $28.50. Laboratory Laboratory costs were very e a s i l y i d e n t i f i e d , given that the labora-tory units entered on the H.S.I form accurately r e f l e c t e d the workload of t h i s department for 1975. The t o t a l number of laboratory units across a l l d i v i s i o n s equalled 29,020,595; the step-down budget of t h i s department, from Table D, i s $7,467,466.10.producing a laboratory u n i t cost of $0,257. But i n f e c t i o n c o n t r o l , research and q u a l i t y c o n t r o l procedures were not removed from the t o t a l . If they were removed, the cost per unit would increase to $0.34 and the r e s u l t i n Chapter IV would have shown a marginal increase i n the d i f f e r -ence of step-down costing methodology from the Per Diem costing method. Each patient's record was reviewed to determine the type and number of laboratory tests performed; standard laboratory units for each test - 70 -have been determined which f a c i l i t a t e the p r i c i n g of each t e s t . The fol l o w i n g examples of major t e s t s f o r each of the major d i v i s i o n s of the laboratory i l l u s t r a t e the degree of s p e c i f i c i t y i n the process of costing used f o r th i s department. Laboratory D i v i s i o n Test T o t a l Lab Units Unit P r i c e Chemistry • glucose 2 $0,257 • sodium 2 0.257 Hematology • CBC 3 0.257 • d i f f e r e n t i a l 8 0.257 Blood Bank • grouping 9 0.257 e cross-match 13 0.257 Histology • s l i d e 10 0.257 • cytology smear 3 0.257 • cytology screen 0.257 (gyn) 5 Isotopes • thyroxine 10 0.257 • f o l a t e 12 0.257 Microbiology • aerobic 3 0.257 • gram 3 0.257 Immunology • pregnancy 2 0.257 • ANA 50 $0,257 T o t a l P r i c e Per Test $ 0.51 0.51 0.77 2.04 2.30 3.34 2.57 0. 77 1.28 2.57 3.08 0.77 0.77 0.51 $12.85 Radiology Radiology offered a large number of tests which were not conducive to grouping. Further, the department did not use a u n i t system comparable to that used i n Laboratory. But the radiology department did have a predetermined fee schedule f o r the physicians s i m i l a r to - 71 -that used i n the EEG department to determine u n i t costs. For the over 50 te s t s , too great a number to l i s t here, the same process was applied to determine i n d i v i d u a l t e s t costs as was used f or the s i x tests of the EEG department. The most frequently used r a d i o l o g i c a l tests by MVA victims are presented here as examples of the unit cost determination. Procedure Rank Based Fee Schedule Rank Number Percentage of Department's Resources New Rate Chest $12.24 S k u l l 18.47 Bone Extremities 12.24 C e r v i c a l Spine 14.73 Voiding Cystogram 30.59 24,496 6,390 27,842 2,588 588 20.30% 6.87% 19.83% 2.22% 1.05% $ 23.95 36.16 23.95 28.59 60.05 T h i r t y per cent of this department's budget i s a l l o c a t e d to medical s t a f f fees. Without a d e t a i l e d study on the a l l o c a t i o n of department resources f o r each te s t provided, i t was f e l t that, f o r t h i s t h e s i s , the fee schedule was the most appropriate method of ranking the component tests of t h i s department. (See s e c t i o n on EEG). The radiology f i l m i s costed by using the same d o l l a r values that the h o s p i t a l paid f o r i t s f i l m . Each medical record shows the type and number of r a d i o l o g i c a l f i l m used; the p r i c e l i s t i s as follows: - 72 -F i l e Type P r i c e B $0.18 C 0.25 D 0.37 E 0.47 F 0.72 G 0.37 Dental 0.31 Pharmacy The t o t a l drug costs, as mentioned e a r l i e r , were a l l o c a t e d across the various service departments using drugs, e.g., ward stock to nursing u n i t s . The drug cost a l l o c a t e d to pharmacy represented the costs of those drugs used i n completing p r e s c r i p t i o n s f o r pat i e n t s . The t o t a l pharmacy costs from Table D were divided i n t o drug costs and non-drug costs. From the medication chart i n the medical records, a complete l i s t of those drugs given to the patients f o r p r e s c r i p t i o n s was determined. With the p r i c e l i s t f o r drugs i n 1975, and the number of doses f o r each drug, a t o t a l d o l l a r value was determined f o r drug cost f o r each pati e n t . Therefore the a l l o c a t e d t o t a l cost from Table D f o r Pharmacy should have drug cost separated and not included i n the following determination of p r e s c r i p t i o n costs. - 73 -Table D A l l o c a t e d Cost f o r Pharmacy - Drug Cost = Pharmacy Production Cost $1,910,505.40 - $1,125,825.34 = $784,680.10 To produce a un i t cost f o r each p r e s c r i p t i o n (not incl u d i n g drug costs) an assumption was made that a l l p r e s c r i p t i o n s would take the same amount of time i n preparation. With t h i s assumption, and with a t o t a l number of p r e s c r i p t i o n s f i l l e d at V.G.H. i n 1975 of 154,098, the u n i t cost f o r f i l l i n g p r e s c r i p t i o n s was $5.09. This unit cost includes overhead to operate the pharmacy and the con s u l t a t i o n time of department personnel with other h o s p i t a l and medical s t a f f . From the medical records, the number of p r e s c r i p t i o n s f i l l e d f o r each patient was determined; t h i s was done by the a s s i s t a n t d i r e c t o r of pharmacy at Vancouver General H o s p i t a l . P h y s i c a l Medicine P h y s i c a l medicine department costs are labour intensive, as can be seen i n Table D. Even though there i s a high overhead due to the large square footage used by th i s department, the amount of time the patient i s i n the department r e f l e c t s the amount of time a.:.' s t a f f member i s with that p a t i e n t . Therefore, the unit value f o r costs f o r p h y s i c a l medicine was determined on an hourly b a s i s . The t o t a l time worked i n p h y s i c a l medicine, as presented on the H.S.2 return form, was 180,758 hours. This hour f i g u r e divided into the - 74 -cost from Table D of $1,771,243 gives the cost f o r patient services o f f e r e d by the p h y s i c a l medicine department of $9.80 per hour.. Time units by the p h y s i c a l medicine department were submitted on each patient's record i n five-minute segments. Therefore, each f i v e -minute segment i n 1975 cost .820. Since there i s no record to determine the services to patients by l e v e l of physiotherapists, occupational t h e r a p i s t s , and speech th e r a p i s t s , i t was f e l t that the assumption of average sa l a r y would s u f f i c e f or the determination of u n i t cost. The research study, therefore, assumed that a l l such s t a f f have equal s a l a r i e s . The remaining d i r e c t p atient services have l i t t l e or no input to the h o s p i t a l care given to MVA v i c t i m s . Only audiology and services department provided services to MVA victims; however, because of the low occurrence of input from these two departments, i t was decided that an average value per audiology test and per hour of s o c i a l s e r v i c e time would be a f a i r i n d i c a t o r of the unit costs of these departments. In the s o c i a l s e r v i c e department, a work u n i t time study completed on a sample of 269 opened and closed cases showed 35% were seen between 12 and 18 minutes. The minimum time was 6 minutes and maximum was 45 minutes. Also, 72.6% of the 4 cases were seen i n two or l e s s contacts. The medical record was quite vague on d e l i n e a t i n g the scope of s o c i a l s ervice v i s i t s to L e t t e r from R.J. Marcus, D i r e c t o r , S o c i a l Services Department, Vancouver General H o s p i t a l , November 4, 1977. - 75 -p a t i e n t s . One-half hour of the s o c i a l worker's time was a l l o c a t e d to each MVA v i c t i m who received services from the department. The t o t a l budget of t h i s department was $746,662, and the t o t a l paid hours from the H.S.2 form was 80,628, which gives an hourly rate of $9.26. The cost of one-half hour was therefore $4.63. The one audiology t e s t noted from the 202 records was costed at $13.85. The t o t a l budget was $88,315 from Table D and the number of t o t a l t e s t s was 6,375. The departments of the H o s p i t a l that did not show i n the analysis of the sample of 202 MVA v i c t i m s ' medical records were r e n a l and psychology. Their costs as l i s t e d i n Table D were not used i n t h i s study. This i s not to say that i n some s p e c i f i c instance these departments would not have input i n t o the treatment of MVA v i c t i m s , but i t would occur r a r e l y . I f necessary, the same procedure for i d e n t i f y i n g the u n i t cost per department could be used and a cost per treatment could be a l l o c a t e d to the patient's t o t a l cost. The f i n a l department to be considered was education. A legitimate question to be r a i s e d at t h i s point i s whether the cost of education i n a h o s p i t a l should be charged to the h o s p i t a l ' s patients or to the community. The cost includes not only s a l a r i e s and r e l a t e d expenses fo r medical students, a l l i e d health students, i . e . , radiology and - 76 -laboratory technicians; i t also includes the i n - s e r v i c e education programs of the h o s p i t a l s t a f f . Some consider t h i s cost to be beyond the d i r e c t cost of providing care to the patient"', but since i t i s a component of both medical t r a i n i n g and continuing education, i t was included i n the costs of treatment. Education costs were d i s t r i b u t e d across a l l patient care areas i n the same manner as general administration (see Table D). Magraw R.W.,M.D., "How Trends i n Medical Education are A f f e c t i n g Medicine and H o s p i t a l , " H o s p i t a l s , October 1, 1963. - 77 -IV. DATA ANALYSIS The previous chapter o u t l i n e d a step-down approach to determining the cost of each s e r v i c e of the H o s p i t a l used by the MVA v i c t i m . Each patient's record was reviewed by the author to determine the number of units i n each patient care area of the h o s p i t a l used by the v i c t i m . When these component costs were t o t a l l e d , an episodic cost was a v a i l a b l e f o r a n a l y s i s of the i n t e r a c t i o n between cost and s e v e r i t y . This cost was l a b e l l e d Step-Down Episodic Cost. A second cost was based on a c a l -culated per diem cost valued at $129.75 (calculated using t o t a l patient days - 605,614 days and t o t a l budget d o l l a r s minus those d o l l a r costs excluded from the step-down c a l c u l a t i o n , i . e . , employee meals and Grandview Nursing Home - $78,578,420). This method of determining an episodic cost represented the t r a d i t i o n a l approach of m u l t i p l y i n g a constant per diem rate times length of stay of the patient's v i s i t i n the h o s p i t a l . For the remainder of the a n a l y s i s , t h i s second cost was r e f e r r e d to as Per Diem Episodic Cost. Two objectives f o r t h i s thesis were i d e n t i f i e d i n Chapter I: • Measure the c o r r e l a t i o n between episodic cost and i n j u r y s e v e r i t y ; and • Determine i f the chosen Step-Down patient s p e c i f i c costing methodology resul t e d i n a s i g n i f i c a n t d i f f e r e n c e from the t r a d i t i o n a l Per Diem Episodic Costing. - 78 -A. Per Diem versus Step-Down Costing i . E pisodic Costs The second o b j e c t i v e l i s t e d e a r l i e r was reviewed f i r s t as a basis f o r b e t t e r understanding of the r e s u l t s of the c o r r e l a t i o n a n a l y s i s . A paired t - t e s t , comparing the Step-Down and Per Diem Episodic Costs f o r each of the 202 patients i n the sample population, was the f i r s t t e s t used to determine i f there was a s i g n i f i c a n t d i f f e r e n c e between the two cost methodologies. TABLE E Paired t-Test Episodic Costs Standard Standard Mean Deviation Error t-Value Per Diem Ep i s o d i c Cost $1,984.15 $2,989.22 $210.32 Step-Down Ep i s o d i c Cost 1,936.10 2,692.86 189.47 0.49 Difference 48.06 1,383.73 97.36 With 201 degrees of freedom, i t was not p o s s i b l e to suggest that there was a s i g n i f i c a n t d i f f e r e n c e between these two groups. The standard d e v i a t i o n of the d i f f e r e n c e was too great compared to the mean of the d i f f e r e n c e . I t was concluded that the t - t e s t was weakened because of the wide range of patients i n t h i s sample which a f f e c t e d the homogeneity neces-sary f o r a t - t e s t , r e s u l t i n g i n t h i s large standard deviation of the - 79 -d i f f e r e n c e . Patients of a l l ages and, more importantly, of a l l s e v e r i t y l e v e l s , were included i n these 202 MVA v i c t i m s . Two d i f f e r e n t approaches were attempted to counteract the influence of the wide range i n s e v e r i t y : • grouping patients by t h e i r A.I.S. score; and • grouping patients i n t o low, medium, and high I.S.S. score categories. Low s e v e r i t y was i d e n t i f i e d f or any patient with an I.S.S. value of 9 or l e s s . This group included those patients who had one i n j u r y of the value 3 i n the A.I.S., or m u l t i p l e i n j u r i e s with a value of 2, 2 and 1 for the top three i n j u r y s i t e s . In t h i s case, the i n j u r i e s were not l i f e - t h r e a t e n i n g . High s e v e r i t y was i d e n t i f i e d f or those pa t i e n t s with a s i n g l e l i f e - t h r e a t e n i n g i n j u r y , i . e . , A.I.S. value of 5, or a combination of i n j u r i e s that gave a comparable I.S.S. score to that of a s i n g l e l i f e - t h r e a t e n i n g i n j u r y . This group included a l l patients with an I.S.S. scale of greater than 24. Medium s e v e r i t y was i d e n t i f i e d as patients having I.S.S. scores between and i n c l u d i n g 10 and 24. Grouping patients by t h e i r A.I.S. score was f e l t to be of value, as t h i s score could e a s i l y be q u a n t i f i e d without accessing patient records. This would be accomplished by using the mapping technique explained e a r l i e r , which tr a n s l a t e s the patient's I.C.D.-9 discharge code. The f o l l o w i n g table (Table F) d i s p l a y s the r e s u l t s of a two-factor (A.I.S. and cost, both Per Diem and Step-Down) ANOVA t e s t , with repeated measures on one f a c t o r ( c o s t ) . - 80 -As can be seen i n Table F, the very low value, 0.0049, f o r the p r o b a b i l i t y that two costing methodologies are s i m i l a r , was s u f f i c i e n t evidence to ind i c a t e a d i f f e r e n c e between the costing methodologies. This same t e s t , a two-factor (I.S.S. and cost) ANOVA with repeated measures on one-factor (cost) , was performed on the two costing methodologies holding the three categories of I.S.S., i . e . , low, medium and high as the independent v a r i a b l e . Table G displays the r e s u l t s of t h i s t e s t . As with the tes t using the components of A.I.S., the r e s u l t s of t h i s two-factor ANOVA t e s t , p=0.118, indicated that there was s i g n i f i c a n t d i f f e r e n c e between the costing methodologies. The r e s u l t s of these two ANOVA tes t s as displayed i n Tables F and G suggested that there was a s i g n i f i c a n t d i f f e r e n c e between the two costing methodologies for episodic costs when the independent v a r i a b l e of s e v e r i t y was categorized into component c e l l s . In f a c t , the prob-a b i l i t y of the two methodologies producing s i m i l a r costs was smaller when s e v e r i t y was categorized by A.I.S. groupings. But for each te s t the standard deviations of the dependent v a r i a b l e s , i . e . , Per Diem and Step-Down episodic costs, were very high f o r most of the c e l l s of both A.I.S. and I.S.S. when compared to the c e l l mean. I f the standard d e v i a t i o n could be reduced, there could be an increase i n the v a l i d i t y of these t e s t s . - 81 -TABLE F .Two-Factor ANOVA with Repeated Measures, Epi s o d i c Cost and A.I.S.  A.I.S. Values 1 2 3 4 5 Marginal Mean 557.4 807.6 2632. 6 2252.9 3875. 2 1984.1 Per Diem Ep i s o d i c Standard Deviation 452.1 872.9 3512. 8 4078.5 3239. 2 Cost Count 27 49 100 11 15 202 Mean 536.4 868.2 2197. 7 2016.9 6140. 6 1936.1 Step-Down Epi s o d i c S tandard Deviation 254.0 1479.8 2516. 5 2896.2 4301. 7 Cost Count 27 49 100 11 15 202 Marginal 546.9 837.9 2415. 2 2134.9 5007. 9 1960.1 Analysis of Variance Across Severity (A.I.S.) F^ d f ^ = 10.84 p = 0.000 Across E p i s o d i c Cost F^ d f 1 = 8.10 p = 0.0049 In t e r a c t i o n Term ¥ = 16.42 p = 0.000 (d.f.4,197) F - 82 -TABLE G Two-Factor ANOVA with Repeated Measures Epi s o d i c Cost and X.S,S. I.S.S. C e l l s Low Severity Medium Severity High Severity Marginal Mean 1409.1 2417.8 4442.4 1984.1 Per Diem Episodic Standard Deviation 2331.3 4686.8 Cost Count 129 52 21 202 Mean 1223.1 2140.8 5809.6 1936.1 T o t a l Episodic Standard 1 9 6 2 < 5 Deviation 1679.4 4678.9 Cost Count 129 52 21 202 Margina1 1316.1 2279.3 5126.0 1960.1 Analysis of Variance Across Severity (I.S.S.) F(d.f.2,199) = 21.24 p = 0.000 Across Episodic Cost F(d.f.1,199) = 6.46 p = 0.0118 In t e r a c t i o n Term F(d.f.2,199) =13.93 p = 0.000 - 83 -The only way to reduce the standard d e v i a t i o n was to lower the range of the d i f f e r e n c e between the Per Diem methodology of costing and the Step-Down methodology of costing. I t was f e l t that the measure f o r the average d a i l y cost for each of these cost methodologies would be the appropriate v a r i a b l e to reduce the range of the d i f f e r e n c e v a r i a b l e and thus reduce the standard deviation. D a i l y Cost In the episodic cost methodology, the length of stay (L.O.S.) of the patient's v i s i t to the h o s p i t a l was an uncontrolled v a r i a b l e . Two d i f f e r e n t p a t i e n t s , having approximately the same Step-Down episodic cost, could have had s i g n i f i c a n t l y d i f f e r e n t lengths of stay. This was not the case for Per Diem episodic cost, as t h i s episodic cost was a d i r e c t m u l t i p l e of L.O.S. and v a r i e d as the L.O.S. v a r i e d . To eliminate the influence of L.O.S. as a way of reducing the standard d e v i a t i o n of the cost v a r i a b l e , the paired t - t e s t and two-factor ANOVA test.with repeated measures were run, using the d a i l y cost of each methodology. Of course, the d a i l y cost of the Per Diem methodology was a constant, but a paired t - t e s t could be used, as the d i f f e r e n c e between the two costs was as randomly d i s t r i b u t e d as was the Step-Down d a i l y cost. Table H displays the r e s u l t s of t h i s paired t - t e s t . - 84 -TABLE H Paired t-Test Da i l y Cost Mean Standard Deviation Standard Err o r t-Value Per Diem Daily Cost Step-Down Dai l y Cost 129.75 0 0 173.33 131.20 9.23 4.72 Dif f e r e n c e 43.58 131.20 9.23 The t-value shown i n Table H suggested that there was a s i g n i f i c a n t d i f f e r e n c e between the two d a i l y rates of each of the costing method-o l o g i e s . What was more important was the r e l a t i v e l y small value of the standard d e v i a t i o n i n t h i s t e s t compared with the standard devi-a t i o n of the episodic costs displayed i n Table E, page 78. The paired t - t e s t of the episodic cost had a mean d i f f e r e n c e of $48.06, while the mean d i f f e r e n c e between the two d a i l y rates was $43.58. While these means were comparatively close, the standard deviations were remarkably d i f f e r e n t , with the d a i l y cost standard deviation ($131.20) being one-tenth of the standard deviation of the episodic cost ($1383.73). Two-factor ANOVA t e s t s , with repeated measures f o r the d a i l y cost, pro-duced high F values of 35.46, p=0.00, when s e v e r i t y was measured i n the three groupings of I.S.S. (see Table K) and 30.33, p=0.00, when s e v e r i t y was categorized by A.I.S. (see Table L ) . Both indi c a t e d a s i g n i f i c a n t d i f f e r e n c e i n the two methodologies when d a i l y costs were compared. In both tables, Tables K and L, the values f o r the s e v e r i t y and the i n t e r -a c t i o n terms were exactly the same, due to the constant value of the d a i l y rate of the Per Diem costing methodology. - 85 -TABLE K Two-Factor ANOVA with Repeated Measures, Dail y Cost and I.S.S.  Per Diem Dail y Cost T o t a l D a i l y Cost Marginal Mean Standard Deviation Count Mean Standard Deviation Count Low Severity 129.75 0.0 129 158.22 86.37 129 143.98 I.S.S. C e l l s  Medium Severity High Severity Marginal 129.75 129.75 0.0 52 173.65 177.71 52 151.70 129.75 0.0 21 265.32 185.33 21 197.53 202 173.32 202 151.54 Analysis of Variance Across Severity (I.S.S.) F ( d . f .2,199) 6.34 P = 0.002 Across Da i l y Cost F ( d . f .1,199) = 35.46 P = 0.000 In t e r a c t i o n Term F ( d . f .2,199) 6.34 P = 0.002 - 86 -TABLE L Two-Factor ANOVA with Repeated Measures, Dai l y Cost and A.I.S.  A.I.S. Values 1 2 3 4 5 Marginal Mean 129.75 129.75 129.75 129.75 129.75 129.75 Per Diem Standard 0.0 0.0 0.0 0.0 0.0 Da i l y Deviation Cost Count 27 49 100 11 15 202 Mean 174.81 162.48 158.72 220.76 268.67 173.33 Step-Down Standard 96.01 96.61 136.97 210.35 142.45 142.45 Dail y Deviation Cost Count 27 49 100 11 15 202 . Marginal 152.28 146.11 144.24 175.25 199.21 151.54 Analysis of Variance Across Severity (A.I.S.) F ( d f 4 197) Across D a i l y Cost F, In t e r a c t i o n Term (d.f.1,197) '(d.f .4,197) 2.83 p = 0.026 30.33 p = 0.000 2.83 p = 0.026 - 87 -FOOTNOTE TO TABLE K AND TABLE L * A comparison of Table K with Table G or Table L with Table F may suggest an apparent anomaly. Table K and Table L show that w i t h i n each ISS or AIS category, the cost per patient-day, c a l c u l a t e d by the step-down method and averaged across a l l patients i n the category, i s very d i f f e r e n t from the hospital-wide per diem. In category 4, Table L, f o r example, the average step-down cost per patient day i s $220.76, much above the hospital-wide per diem of $129.75. On the other hand, the cost per episode, c a l c u l a t e d by the step-down method, i s only $2,016.97, compared with the per diem based episode cost of $2,252.93. Yet average length of stay f o r both groups i s the same, 17.36, because of course they are the same pati e n t s , costed by two d i f f e r e n t methodologies. The explanation i s a strong negative c o r r e l a t i o n between length of stay and step-down based cost per patient day. I f we designate the f u l l y a l l o c a t e d cost of an episode of care for a s p e c i f i c p a tient i , measured i n t h i s study, as E., the corresponding length of stay as D., and the hospital-wide per diem as P, then the average cost per episode f o r a p a r t i c u l a r group of n p a t i e n t s , c a l c u l a t e d by the step-down method, i s E^ = — £ E. , but by the per diem method i s E P = — P V D. = PD. n . - I n . T . x 1=1 . 1=1 We can c a l c u l a t e P. = E. f D. as the cost per day, derived by the step-i i i „ i n „ —s 1 s down method, f o r each patient i , and can then c a l c u l a t e P = — .2 P. , n . , l i = l g g but i t w i l l not be true i n general that f D = E . Indeed, i n Table L, ISS group 4, IT* = 220.76, "5 = 17.36, t h e i r product i s $3,832.39, f a r above E S at $2,016.97. Factoring E , we can write: -s 1 n 1 n s s E = — y E. = — ED..p. , and d e f i n i n g d. and p. as the amount by n . . i n . T l * i & l * i J i = l i = l g which the values of D. and P. f o r patient i deviate from the mean values I l of the group, I s = i £ (D + d.) (P S + p*) i = l , n __ _ _ = - r (DP S + Dp + P Sd. + d.pf) n * i l i * i i = l = DP S + ± d.p 1 n i = l i s The middle terms become zero on summation over the group as a whole. - 88 -FOOTNOTE TO TABLE K AND TABLE L (Continued) n s s I f D. and P. are uncorrelated, then V d.p.—>0 i n the l i m i t as 1 x i * i ^ n—> o o though f o r an a c t u a l set of observations on small n i t w i l l not. g In t h i s case, though, D_^  and P^ turn out to be strongly negatively c o r r e l a t e d - some patients receive very i n t e n s i v e care a short time, others i n the same s e v e r i t y group may survive a very long time at lower n l e v e l s of care. So T, d.p. i s large and negative. i = l 1 1 This observation underscores the importance of focussing on the o v e r a l l episode of care, rather than the per diem costs, i n making comparisons among patients with d i f f e r e n t types of problems. I t i n d i c a t e s , however, that while s e v e r i t y measures explain a s u b s t a n t i a l amount of the v a r i a -t i o n i n costs, they do not by any means do the whole job, and considerable v a r i a t i o n remains. A l l of t h i s , however, i s masked by t r a d i t i o n a l l y focussing only on hospital-wide per diems and patient lengths of stay. - 89 -B. Analysis of the Influence of Severity on the V a r i a t i o n i n Cost. The primary o b j e c t i v e of t h i s thesis has been to i d e n t i f y the r e l a t i o n -ship between cost and s e v e r i t y , and determine i f the Step-Down method of costing was more responsive to s e v e r i t y than was the Per Diem method of costing. The f i r s t task i n t h i s s e c t i o n of data analysis was the mul t i p l e regression of episodic costs as the dependent v a r i a b l e , and se v e r i t y as the independent v a r i a b l e . Table M displays the regression of the episodic costs f o r Step-Down and Per Diem methods where s e v e r i t y was measured i n I.S.S. values. Table N displays the regression of the episodic costs f o r Step-Down and Per Diem methods where s e v e r i t y was measured i n A.I.S. scores. TABLE M Regression Episodic Costs With I.S.S. Per Diem Step-Down M u l t i p l e R 0.335 0.507 R 2 0.112 0.257 2 Adjusted R 0.108 0.253 F-value 25.22 69.02 - 90 -TABLE N Regression Episodic Costs With A.I.S. Per Diem Step-Down M u l t i p l e R 0.317 0.462 0.101 0.213 Adjusted R' 2 0.096 0.209 F-value 22.39 54.25 For a l l regressions of Per Diem and Step-Down episodic costs with s e v e r i t y , the F-values showed s i g n i f i c a n c e . Therefore, s e v e r i t y does have an impact on cost; but the r e a l value of t h i s t e s t was the quanti-f i c a t i o n of the explanatory power that s e v e r i t y had with episodic costs. 2 The adjusted R value f o r Step-Down costs and I.S.S., 0.253, suggested that 25.3% of the v a r i a t i o n i n the Step-Down costing methodology was a t t r i b u t a b l e to, or explained by, co-variance with the s e v e r i t y as 2 measured by the I.S.S. In comparison, the adjusted R value f o r the regression of Per Diem episodic costs and I.S.S., 0.108, suggested only 10.8% of the v a r i a t i o n i n Per Diem episodic costs was a t t r i b u t a b l e to co-variance with s e v e r i t y as measured by I.S.S. Quite c l e a r l y , the Step-Down method of costing f o r episodic costs was more responsive to s e v e r i t y than was Per Diem methodology when s e v e r i t y was measured by I.S.S. - 91 -For the two regressions of Per Diem and Step-Down episodic costs with s e v e r i t y measured i n A.I.S. scores, the F-values showed s i g n i f i c a n c e . Again, the value of these equations was t h e i r a b i l i t y to quantify the 2 explanatory power A.I.S. had with episodic costs. The adjusted R value f o r Step-Down and A.I.S., 0.209, suggested that 20.9% of the v a r i a t i o n i n t h i s cost methodology was a t t r i b u t a b l e to, or explained by, co-variance 2 with A.I.S. In comparison, the adjusted R value for the regression of Per Diem episodic costs and A.I.S., 0.096, suggested that only 9.6% of the v a r i a t i o n i n Per Diem cos t i n g was a t t r i b u t a b l e to co-variance with s e v e r i t y as measured by A.I.S. The A.I.S. scores did not seem to have as high an explanatory power over the v a r i a t i o n of episodic costs as did the I.S.S. scores. To further analyze t h i s r e l a t i o n s h i p between episodic costs and s e v e r i t y , regression analysis was performed on each of the Step-Down cost components, as defined i n Chapter I I I of t h i s t h e s i s , and s e v e r i t y measured by both I.S.S. and A.I.S. Table P presents the r e s u l t s of regressions of component costs with s e v e r i t y . Only the laboratory cost 2 2 had a higher adjusted R value than the Step-Down episodic cost per R value. - 92 -TABLE P Regression of Component Costs With Severity Component Cost I.S.S Severity A.I.S 2 Adjusted R F-Value 2 Adjusted R F-Value Ward Cost 0.207 53.5 0.169 41.8 Operating Room Cost 0.115 27.08 0.120 28.5 Radiology 0.149 36.15 0.086 20.0 Laboratory 0.307 89.85 0.255 69.7 Drugs/Pharmacy 0.039 9.09 0.031 7.4 Physiotherapy 0.058 13.43 0.046 10.6 There was l i t t l e value i n furt h e r exploring the r e l a t i o n s h i p of these component costs; the focus of t h i s thesis was on the episodic costs or i t s d a i l y r a t e s . Table P was presented to the reader to i d e n t i f y that no one component co n t r o l l e d the variance of Step-Down episodic costs i n i t s r e l a t i o n s h i p with s e v e r i t y . - 93 -Test f o r C u r v i l i n e a r i t y To ensure that l i n e a r regression was the best approach f o r measuring the r e l a t i o n s h i p between episodic costs and s e v e r i t y , the following t e s t f o r c u r v i l i n e a r i t y was performed. For t h i s a p p l i c a t i o n , the i n t e r -v a l p r e d i c t o r v a r i a b l e with d i s c r e e t values was d i r e c t l y represented by dummy v a r i a b l e s . The s i g n i f i c a n c e of the d e v i a t i o n from l i n e a r i t y was tested by comparing the r e s u l t s of l i n e a r regression with the r e s u l t s of dummy regression. In t h i s case, the independent v a r i a b l e , s e v e r i t y measured by I.S.S. and A.I.S., was entered i n t o the regression a n a l y s i s , using dummy v a r i a b l e s f o r the components of the I.S.S. and A.I.S. s e v e r i t y s c a l e s . The relevant F-test was: 2 2 (R with Dummy Variables-R with o r i g i n a l v a r i a b l e s ) / ( k ) F = : (1-R with Dummy Variables)/(N-k-1) N = sample s i z e k = number of dummy va r i a b l e s 2 For the four tests f o r c u r v i l i n e a r i t y , that i s , comparing the R value of the r e l a t i o n s h i p of both episodic costs with I.S.S. and the Revalue 2 of both episodic costs with A.I.S., the relevant R values and the r e s u l t i n g F-test value are displayed i n the following table, TableQ . TABLE Q Test f o r C u r v i l i n e a r i t y 2 2 R with R with Degrees of Dummy Varia b l e s O r i g i n a l V a r i a b l e s F-Value Freedom Step-Down with I.S.S. 0.262 0.257 0.67 1 and 199 Step-Down with A.I.S. 0.261 0.213 3.2 3 and 197 Per Diem with I.S.S. 0.100 0.112 1.33 1 and 199 Per Diem with A.I.S. 0.122 0.101 1.18 3 and 197 - 94 -Only the F-value f o r Step-Down with A.I.S. regression showed s i g n i f i c a n c e at p=0.05, but not at p=0.01. Considering that A.I.S. had only f i v e cate-g o r i e s , i t was f e l t that these r e s u l t s did not s u f f i c i e n t l y r e j e c t 1: i t y f o r regressions of cost with s e v e r i t y . Therefore, i n the remaining a n a l y s i s , s t r a i g h t l i n e l i n e a r regression was used. .xnear-Influence of Other Independent Variables Other v a r i a b l e s s p e c i f i c to each patient d i d modify the r e l a t i o n s h i p between episodic cost and s e v e r i t y . As seen e a r l i e r , length of stay was one v a r i a b l e which had s i g n i f i c a n t impact on the r e l a t i o n s h i p between cost and s e v e r i t y . This v a r i a b l e was already included i n the d e f i n i t i o n of episodic cost. But f a c t o r s such as age of the patient, sex of the patient, whether the patient had an operation, whether the patient died, and the l o c a t i o n of the i n j u r y - were a l l f a c t o r s that could have i n f l u -enced the r e l a t i o n s h i p between episodic cost and s e v e r i t y , and that were measured during the a n a l y s i s of the p a t i e n t s ' medical records. The following graph, Graph 1, displays the d i s t r i b u t i o n of patients' ages by the s e v e r i t y of t h e i r i n j u r y , measured by I.S.S. With the sampling process based on sex and s e v e r i t y , there was p o t e n t i a l to have poor representation of one or more age groups. But only the e l d e r l y with high s e v e r i t y seem to have been poorly represented i n the sample. Pos s i b l y they d i d not e x i s t i n the population, on the assumption that the e l d e r l y would not be able to survive a high s e v e r i t y i n j u r y as w e l l as a younger person. - 95 -In the following m u l t i p l e regression, episodic cost remained as the dependent v a r i a b l e while the independent v a r i a b l e included those items l i s t e d above as w e l l as s e v e r i t y . This equation was processed, using S.P.S.S. with a step-wise methodology that allowed c o n t r o l over the entry of the independent v a r i a b l e s , f o r both Step-Down and Per Diem episodic costs f o r each of the s e v e r i t y s c a l e s . The r e s u l t s follow i n Tables R and S for s e v e r i t y measured by I.S.S. and Tables T and U for s e v e r i t y measured by A.I.S. - 96 -GRAPH 1  Severity By Age * # » * * * # * * » « * * * • * * * o n (NJ * o * * Cvl « * * * * o « # 1 •» i. « in •J) • * * * * * o i in . • * 1 '0 * » * * * * o * * * • * • CVJ * « • o * « N « • • * * * » ?l N * « * * * * * * o * n* f> * * 3 » CM N CJ -• • n * . <r * « CJ * « * CM * • « CM • o • • * * o * « * * * « o o o c o O O 10 o o o n o I.S.S. - 97 -TABLE R Regression of Episodic Cost With Multiple Independent Variables and I.S.S, Per Diem Cost Step-Down Cost Severity Entered on F i r s t Step i n Dur. y Format: Adjusted R 2 = 0.091 Adjusted R 2 = 0.256 F-value = 11.04 F-value = 35.45 Age Entered on Second Step i n Dummy Format: Adjusted R 2 = 0.100 ' Adjusted R 2 = 0.260 F-value = 4.71 F-value = 12.78 Sex Entered on Third Step in Dummy Format: Adjusted R 2 = 0.095 Adjusted R 2 = 0.256 F-value = 4 - 0 2 F-value = 1 0 - 9 2 Operation not Performed Entered on Fourth Step in Dummy Format: Adjusted R 2 = 0.209 Adjusted R2 = 0.370 F-value =7.65 F-value = 15.77 Death as an Outcome Entered on F i f t h Step in Dummy Format: 2 9 Adjusted R = 0.246 Adjusted R = °-43 F-value = 8.3 F-value = 18.13 Injury Location Entered on Sixth Step i n Dummy Format: 2 9 Adjusted R = 0.271 Adjusted R = 0.443 F-value = 6.34 F-value = 12.46 Severity 2nd order Interactive Terms Entered on Seventh Step: Adjusted R2 = 0.270 Adjusted R2 = 0.468 F-value = 5.14 F-value = 10.83 Age 2nd order Interactive Terms Entered on Eighth Step: 2 9 Adjusted R = 0.272 Adjusted R = 0.477 F-value =4.59 F-value = 9.72 Sex, Operation and Death 2nd order Interactive Terms Entered on Ninth S' Adjusted R 2 = 0.263 Adjusted R 2 = 0.483 F-value = 3.99 F-value = 8-82 Injury Location 2nd order Interaction Terms Entered on Tenth Step: Adjusted R 2 = 0.25 Adjusted R 2 = 0.479 F-value = 3.31 F-value = 7.39 - 98 -TABLE S C o - e f f i c i e n t s of Independent Variables and I.S.S. Step-Down Episodic Cost and Severity Measured by I.S.S, VARIABLE B I.S.S. I.S.S. Age Age Age Injury Injury Injury Injury Injury 0- 9 10-24 0-14 15-44 45-64 Sex (Female) Operation Not Performed Death as Outcome Head Face Chest Abdomen Extremities Sex x Severity Operation x Severity Death x Severity Age x Severity Age x Sex Age x Operation Age x Death Sex x Operation Sex x Death Operation x Death Severity x Location Age x Location Sex x Location Operation x Location Death x Location Constant -5189.45 -5010.14 -2254.44 -1591.42 -728.03 592.10 -2025.32 13348.87 1186.69 850.23 1802.68 -377.21 1615.18 -4.97 160.95 -470.15 0.43 -1.04 -23.83 -41.18 -939.84 -3483.28 -3149.97 0.000 0.000 0.000 0.000 0.000 7 7 4 8 . 5 6 BETA STD ERROR B F -0.92 895.92 33.55 -0.81 783.88 40.85 -0.26 1108.12 4.13 -0.28 962.27 2.73 -0.09 992.63 0.53 o.io 968.52 0.37 -0.37 794.58 6.49 0.90 6723.79 3.94 0.19 1389.34 0.73 0.08 1404.41 0.36 0.19 1449.66 1.54 -0.04 1487.82 0.06 0.28 1361.37 1.40 -0.01 43.60 0.01 0.39 47.72 11.37 -0.94 147.15 10.20 0.00 0.86 0.00 -0.00 15.18 0.00 -0.20 15.26 2.43 -0.10 44.79 0.84 -0.15 709.55 1.75 -0.12 2292.75 2.30 -0.16 1905.01 2.73 0.05 0.00 0.38 -0.02 0.00 0.07 -0.03 0.00 0.26 -0.24 0.00 2.11 -0 14 0.00 0.79 - 99 -TABLE S (Cont'd.) 11. Per Diem Episodic Cost and Severity Measured by I.S.S, VARIABLE I.S.S. 0- 9 I.S.S. 10-24 Age 0-14 Age 15-44 Age 45-64 Sex (Female) Operation Not Performed Death as Outcome Injury - Head Injury - Face Injury - Chest Injury - Abdomen Injury - Extremities Sex x Severity Operation x Severity Death x Severity Age x Severity Age x Sex Age x Operation Age x Death Sex x Operation Sex x Death Operation x Death Severity x Location Age x Location Sex x Location Operation x Location Death x Location Constant B -3921.08 -3506.14 -1067.14 -399.51 420.30 119.83 -1612.52 5762.51 1800.42 1251.79 1935.37 1394.01 2942.66 -32.64 113.66 -283.70 -0.52 15.00 -27.27 -15.51 -598.82 -1445.36 -701.08 0.000 0.000 0.000 0.000 0.000 4704.74 BETA -0.63 -0.51 -0.11 -0.06 0.05 0.01 -0.26 0.35 0.26 0.11 0.18 0.14 0.46 -0.07 0.24 -0.51 -0.07 0.11 -0.20 -0.03 -0.08 -0.04 -0.03 0.0 -0.02 -0.02 -0.22 -0.08 STD ERROR B 1193.98 10.78 1044.67 11.26 1476.78 0.52 1282.41 0.09 1322.87 0.10 1290.74 0.00 1058.93 2.31 8960.72 0.41 1851.55 0.94 1871.64 0.44 1931.95 1.00 1982.80 0.49 1814.29 2.63 58.11 0.31 63.60 3.19 196.11 2.09 1.15 0.20 20.23 0.55 20.34 1.79 59.69 0.06 945.61 0.40 3055.52 0.22 2538.79 0.07 0.00 0.00 0.00 0.06 0.00 0.09 0.00 1.31 1094.66 0.18 - 100 -The r e s u l t s displayed i n Tables R and S f o r the regression of episodic cost with s e v e r i t y measured by I.S.S., are self-explanatory. The independent v a r i a b l e s were introduced into the equation i n ten d i f f e r e n t steps, but only three of these produced meaningful changes i n the 2 adjusted R value. These were s e v e r i t y , whether an operation was per-formed, and death as an outcome. Age, sex, i n j u r y l o c a t i o n , and most of 2 the second order i n t e r a c t i o n had l i t t l e impact on the adjusted R value. The F-values i n Table S showed that one age group, that being the f i r s t age group with age valued between 0 and 14, and one body l o c a t i o n , that being the abdomen, might have had some impact on the v a r i a t i o n of episodic cost. But a subsequent regression performed without those independent v a r i a b l e s with low F-values, i n d i c a t e d that the age groups 0-14 and 15-44, and body location-abdomen, did not s i g n i f i c a n t l y a f f e c t the r e l a t i o n s h i p between episodic cost and s e v e r i t y . Therefore, the independent v a r i a b l e s included i n the regression model to determine episodic costs were s e v e r i t y (measured by I.S.S.), operation not performed, death as outcome, and t h e i r i n t e r a c t i o n terms. Tables T and U d i s p l a y the r e s u l t s of regression analyses of episodic costs with the same independent v a r i a b l e s as above, except f o r s e v e r i t y , which 2 was measured by A.I.S. scores. The adjusted R value of Step-Down costs 2 with A.I.S. regression was l a r g e r than the adjusted R value of Per Diem and A.I.S. f o r each of the ten steps. The f i n a l F-value, 2.63, f o r the re-gression of Per Diem episodic cost and the independent v a r i a b l e s , was 101 -TABLE T Regression of E p i s o d i c Cost With M u l t i p l e Independent V a r i a b l e s and A.I.S. Per Diem Cost Step-Down Cost Se v e r i t y Entered on F i r s t Step i n Dummy Format: Adjusted R 2 = 0.104 Adjusted R 2 = 0.246 F-value =6.85 F-value = 17.42 Age Entered on Second Step i n Dummy Format: 2 9 Adjusted R =0.114 Adjusted R = 0.253 F-value = 4.22 F-value = 9.51 Sex Entered on T h i r d Step i n Dummy Format: Adjusted R 2 = 0.109 Adjusted R 2 = 0.249 F-value = 3-?3 F-value = 8.41 Operation Not Performed Entered on Fourth Step i n Dummy Format: Adjusted R 2 = 0.206 Adjusted R 2 = 0.353 F-value =6.23 F-value = 11.97 Death as an Outcome Entered on F i f t h Step i n Dummy Format: Adjusted R 2 = 0.227 Adjusted R 2 = 0.390 F-value = 6 «35 F-value = 12.67 Injury Location Entered on Si x t h Step i n Dummy Format: 2 9 Adjusted R = 0.240 Adjusted R = 0.399 F-value =4.96 F-value = 9.33 Severity 2nd order I n t e r a c t i v e Terms Entered on Seventh Step: Adjusted R 2 = 0.226 Adjusted R 2 = 0.395 F-value =3.94 F-value = 7.56 Age 2nd order I n t e r a c t i v e Terms Entered on Eighth Step: Adjusted R 2 = 0.224 Adjusted R 2 = 0.395 F-value =3.53 F-value = 6.79 Sex, Operation and Death 2nd order I n t e r a c t i v e Terms Entered on Ninth St Adjusted R 2 = 0- 2 1? Adjusted R 2 = 0.393 F-value =3.15 F-value = 6.01 Injury Location 2nd order I n t e r a c t i o n Terms Entered on Tenth Step: Adjusted R 2 =0.202 Adjusted R 2 = 0.388 F-value =2.63 F-value = 5 . 1 0 102 -TABLE U C o - E f f i c i e n t s of Independent Variables and A.I.S, i . Step-Down Episodic Cost and Severity Measured by A.I.S. VARIABLE B A.I.S. - 1 -3352.62 A.I.S. - 2 -5371.66 A.I.S. - 3 -4690.26 A.I.S. - 4 -3687.42 Age 0-14 -1177.91 Age 15-44 -546.35 Age 45-64 -23.96 Sex (Female) 701.98 Operation Net Performed -3085.22 Death as Outcome -20607.62 Injury - Head 3784.85 Injury - Face 3571.03 Injury - Chest 5182.43 Injury - Abdomen 3091.55 Injury - Extremities 4457.86 Sex x Severity -90.05 Operation x Severity 833.85 Death x Severity 2928.78 Age x Severity 3.51 Age x Sex -2.51 Age x Operation -23.24 Age x Death -40.10 Sex x Operation -658.86 Sex x Death -1035.12 Operation x Death -711.06 Severity x Location 0.000 Age x Location 0.000 Sex x Location 0.000 Operation x Location 0.000 Death x Location 0.000 Constant 3537.63 BETA STD ERROR B F -0.42 2022.67 2.74 -0.85 1333.57 16.22 -0.87 987.04 22.58 -0.31 1154.10 10.20 -0.13 1307.09 0.81 -0.09 1074.47 0.25 -0.00 1086.40 0.00 0.12 1548.03 0.20 -0.56 1900.13 2.63 -1.40 46021.24 0.20 0.62 2576.65 2.15 0.35 2680.87 1.77 0.54 2858.16 3.28 0.35 2784.23 1.23 0.77 2650.22 2.82 -0.04 417.79 0.04 0.42 609.44 1.87 0.88 8617.24 0.11 0.08 6.89 0.26 -0.02 16. 70 0.02 -0.19 16.84 1.90 -0.10 47.53 0.71 -0.10 789.33 0.69 -0.03 1998.56 0.26 -0.03 2238.12 0.10 -0.22 0.00 0.88 -0.29 0.00 0.07 -0.21 0.00 0.06 -0.27 0.00 1.73 0.41 3049.30 0.46 -103 -TABLE U (Cont'd.) 1 1 . Per Diem Episodic Cost and Severity Measured by A.I.S VARIABLE A.I.S. - 1 A.I.S. - 2 A.I.S. - 3 A.I.S. - 4 Age 0-14 Age 15-44 Age 45-64 Sex (Female) Operation Not Performed Death as Outcome Injury - Head Injury - Face Injury - Chest Injury - Abdomen Injury - Extremities Sex x Severity Operation x Severity Death x Severity Age x Severity Age x Sex Age x Operation Age x Death Sex x Operation Sex x Death Operation x Death Severity x Location Age x Location Sex x Location Operation x Location Death x Location Constant B -2769.73 -3870.70 -1958.10 -856.62 -211.98 518.31 841.32 855.34 122.55 -26080.24 3202.20 2752.95 4348.06 3355.93 4485.78 -286.20 -305.53 4327.63 1.95 9.85 -23.93 -23.10 -670.26 250.54 2107.93 0.000 0.000 0.000 0.000 0.000 708.76 BETA STD ERROR B F -0.31 2563.71 1.16 -0.55 1690.29 5.24 -0.32 1251.06 2.45 -0.06 1462.81 0.34 -0.02 1656.72 0.01 0.08 1361.88 0.14 0.10 1377.00 0.37 0.13 1962.11 0.19 0.02 2408.39 0.00 -1.59 58331.38 0.20 0.47 3265.88 0.96 0.24 3397.97 0.65 0.41 3622.68 1.44 0.34 3528.97 0.90 0.70 3359.13 1.78 -0.13 529.55 0.29 -0.13 772.45 0.15 1.18 10922.25 0.15 0.04 8.73 0.05 0.07 21.17 0.21 -0.18 21.35 1.25 -0.05 60.25 0.14 -0.09 1000.47 0.44 -0.00 2533.15 0.01 -0.09 2836.79 0.55 -0.27 0.00 1.01 -0.03 0.00 0.08 -0.01 0.00 0.02 -0.10 0.00 0.19 0.24 3864.95 0.12 - 104 -c l o s e to the threshold measure of s i g n i f i c a n c e f o r t h i s F-value, (degrees of freedom 31 and 170 F = 1.46 at p = 0.05). From Table U, i t can be noted that for t h i s regression s e v e r i t y , operation performed, body l o c a t i o n s and several of t h e i r i n t e r a c t i o n terms displayed s i g n i f i -cant F-values. A subsequent regression, without those independent v a r i a b l e s with low F-values, i n d i c a t e d that only the F-value for s e v e r i t y was s i g n i f i c a n t i n the regression of episodic costs, with s e v e r i t y measured by A.I.S. scores. - 105 -Regression Model The objective of t h i s thesis has been to measure the strength of the impact of s e v e r i t y on the measurement of episodic cost. In the regres-s i o n of episodic cost and s e v e r i t y , s e v e r i t y measured by I.S.S. proved to be a better p r e d i c t o r of v a r i a t i o n of episodic cost than s e v e r i t y measured by A.I.S. For that reason, the following regression model was developed s o l e l y f o r s e v e r i t y measured by I.S.S. The f i r s t r e g ression model was b u i l t using cost as the dependent v a r i a b l e and s e v e r i t y measured by I.S.S., operation not performed and death as outcome as the independent v a r i a b l e s . The second model included the i n t e r -a ction terms of operation and death with s e v e r i t y . The following table, Table V, displays the r e s u l t s of those regression analyses. For both costing methodologies, the a d d i t i o n of i n t e r a c t i o n terms seemed 2 to have had l i t t l e e f f e c t on the adjusted R value. But these extra independent v a r i a b l e s did reduce the l e v e l of s i g n i f i c a n c e as measured by 2 the F-term. The minor decrease i n the adjusted R values compared with 2 the adjusted R values displayed e a r l i e r i n Table R, resulted from having fewer independent v a r i a b l e s and having introduced s e v e r i t y i n t o the equation i n metric form, not c a t e g o r i c a l form or dummy form. The r e s u l t i n g equations f o r the f i r s t model are displayed i n Graphs 2 and 3 f o r Step-Down and Per Diem episodic cost, r e s p e c t i v e l y . Having death as an outcome appeared to be an important v a r i a b l e as the equation l i n e s - 106 -f o r non-survivors were s u b s t a n t i a l l y removed from the o r i g i n a l regression l i n e . In both costing methodologies, when a patient died, the model was v a l i d only at high s e v e r i t y and predicted a much lower cost of treatment compared with the cost of treatment of a patient with the same s e v e r i t y who survived. The impact of having an operation was consistent for each costing method-ology and whether the patient survived or died; that i s , the o v e r a l l costs increased when the patient had an operation, compared to the patient with the same s e v e r i t y who did not have an operation. Regardless of whether or not episodic costs were measured by a Per Diem method or by costing a l l services of the h o s p i t a l used by the patient, having an operation had a comparable e f f e c t i n increasing cost. When the i n t e r a c t i o n terms were introduced to the two regression equations (the second model), there were minor changes i n the c o - e f f i c i e n t v a r i a b l e s . But the most s i g n i f i c a n t change occurred i n the p value f o r death as outcome. This was e s p e c i a l l y evident for the Step-Down method where )B changed from -$4,218.16 to +$1,125.73 from the f i r s t model to the second model. Because the i n t e r a c t i o n term of "death x s e v e r i t y " used s e v e r i t y i n the metric form, not i n dummy form, the r e a l s e v e r i t y measure of the v i c t i m was used to c a l c u l a t e the i n t e r a c t i o n term. The following ex-amples give the value of t h i s i n t e r a c t i o n term of two v i c t i m s , one with s e v e r i t y '3' and one with s e v e r i t y '30.' - 107 -TABLE V C o - e f f i c i e n t s of Independent Variables i n the Regression Models  Per Diem Step-Down F i r s t Model(Without In t e r a c t i o n Terms) p Values - I.S.S. 0-9 -2921.58 - I.S.S. 10-24 -2284.87 - Operation Not Performed -1970.11 - Death as Outcome -3599.51 - Constant 5674.68 M u l t i n l e R Adjusted R F-Value 0.503 0.253 0.238 16.72 -4751.03 r4113.62 -1708.22 -4218.16 7139.33 0.665 0.442 0.431 39.06 Second Model(With In t e r a c t i o n Terms) B Values - I.S.S. 0-9 -2664.81 - I.S.S. 10-24 -2495.20 - Operation Not Performed -2777.87 - Death as Outcome - 431.65 - Operation* x Severity 88.43 - Death x Severity - 155.13 - Operation5'5 x Death 432.35 - Constant 5659.78 M u l t i p l e R 0.521 R2 0.271 Adjusted R 0.245 F-Value 10.3 -4510.07 -4446.73 -2642.12 1125.73 110.29 - 210.54 -1151.60 7149.85 0.688 0.473 0.454 24.85 Operation not performed. - 108 -GRAPH 2 - Model 1 Step-Down Costs By Operation, Death and I•S.S. SteD-Down Episodic Costs - 110 -Value of I n t e r a c t i o n Term Death x Severity  I n t e r a c t i o n Term = B x Death* x Severity f o r ISS = 3 = -210.54 x 1 x 3 = - $ 631.62 f o r ISS =30 = -210.54 x 1 x 30 = - $6,316.20 * Death was entered into regression equation i n dummy format value of ' I ' i f death occurred and '0' i f v i c t i m survived. At any value of ISS over 6, the i n t e r a c t i o n term counteracts the p o s i t i v e B f o r death as outcome. I f a low s e v e r i t y v i c t i m dies, i t i s l o g i c a l that more h o s p i t a l resources would be used than f o r low s e v e r i t y patients who survived. Conversely, i f a high s e v e r i t y patient dies, i t i s l o g i c a l that fewer resources are used f o r t h i s patient than f o r a comparable patient who survived. The data support t h i s argument. The i n t e r a c t i o n term "operation x s e v e r i t y " behaves i n a s i m i l a r manner. As s e v e r i t y increases, the regression l i n e i n Graph 4 and Graph 5 f o r "no operations" veers into the "with operation" l i n e . At high s e v e r i t y t h i s i n t e r a c t i o n term counteracts the B value f o r 'operation not performed.' But i n t h i s sample of h o s p i t a l i z e d MVA victims there were v i r t u a l l y no patients with a high s e v e r i t y score who did not have an operation regard-l e s s of whether they survived or died. - I l l -GRAPH 4. - Model 2 Step-Down Costs By Operation, Death, Severity and I n t e r a c t i o n Terms o m » iO r.' f Li N o '33 o •Tl •SI • • • • • • • • » • a • _ N M •SI c rj N c O Z* c ' c in SI o o •e "1 o . •» —i — — ** Step-Down Costs - 112 -GRAPH 5 - Model 2 Per Diem Costs By Operation, Death, Severity and I n t e r a c t i o n Terms  - 113 -The r e a l change with the introduction of the i n t e r a c t i o n terms was the decrease i n slope for patients who died. With the second model the pre-d i c t i v e power f o r those patients who died was minimal, regardless of whether or not the patient had an operation. The same type of change, going from Model 1 to Model 2, occurred with the Per Diem Methodology f o r c o s t i n g (see Graph 5). L i t t l e change was noticed for patients who survived but, f o r those who died, the slope be-came negative, suggesting that as s e v e r i t y increases, cost decreases. While the number of patients who died was quite small and thus d i d not s i g n i f i c a n t l y influence the t o t a l sample, t h i s outcome was not s u r p r i s -ing for t h i s type of costing system. With death as the outcome, the more severe the i n j u r y , the shorter the expected.length of stay. Results from the Two-Factor ANOVA t e s t s , multiple regression with several independent v a r i a b l e s and with selected independent v a r i a b l e s , i d e n t i -f i e d that cost was s e n s i t i v e to s e v e r i t y regardless of whether cost was measured by I.S.S. or A.I.S. Based on the assumption that the costing method most s e n s i t i v e to s e v e r i t y was the best measure of cost, i t was concluded that Step-Down costing was a superior method to Per Diem cost-ing for determining p a t i e n t s ' episodic h o s p i t a l costs. Based on t h i s same l o g i c , i t was concluded that I.S.S. was a better measure of s e v e r i t y than was A.I.S., as both costing methodologies were more s e n s i t i v e to the former measure of s e v e r i t y . - 114 -V. SUMMARY AND CONCLUSIONS Episodic h o s p i t a l costs have been measured i n t h i s analysis to determine t h e i r s e n s i t i v i t y to s e v e r i t y . Using a sample of motor v e h i c l e victims to represent a l l h o s p i t a l p a t i e n t s , i t was shown that s e v e r i t y , when measured by I.S.S., was responsible for up to 46% of the v a r i a t i o n of Step-Down costs under c e r t a i n conditions. This value was 8% higher than the r e l a t i o n s h i p that s e v e r i t y had, when measured by A.I.S. with Step-Down costs. The r e l a t i o n s h i p i s u s e f u l to the extent that i t can be applied i n a n a l y t i c a l studies on h o s p i t a l costs, such as cost con-tainment. Before t h i s can be done, there must be a d d i t i o n a l research to determine i f the c o r r e l a t i o n can be improved by changing or adding v a r i a b l e s i n the regression a n a l y s i s . These changes would focus on the following v a r i a b l e s : • sample population - MVA v i c t i m • i n j u r y s e v e r i t y scale - I.S.S. • Step-Down costing methodology • other independent v a r i a b l e s - a v a i l a b i l i t y of family support or home care - day of admission or discharge. Sample Population - MVA V i c t i m The two reasons f o r choosing t h i s p a r t i c u l a r group of patients were ou t l i n e d e a r l i e r i n t h i s report. These were the p o t e n t i a l return to cost b e n e f i t a n a l y s i s of t r a f f i c accident prevention programs, and the wide range of i n j u r i e s to d i f f e r e n t body locatio n s for otherwise healthy i n d i v i d u a l s . I t was assumed that these trauma patients represented -.115 -a l l trauma vi c t i m s , regardless of the cause of i n j u r y . The average episodic Step-Down cost for these MVA victims was quite close to the average episodic Per Diem cost, suggesting that MVA victims o v e r a l l used se r v i c e s i n a comparable manner to a l l patients of the h o s p i t a l s . The length of stay of MVA v i c t i m s , however, was s i g n i f i c a n t l y d i f f e r e n t from a l l VGH patients separated from the short term u n i t i n 1975. MVA victims have 15.2 days and an average length of stay (with 41% being below 5 days and 21% being above 20 days). The average length of stay for the h o s p i t a l was 10.5 days for short term u n i t p a t i e n t s . Another group of patients (e.g., p l a s t i c surgery, cancer care, e l e c t i v e surgery) should be used to repeat t h i s study to determine i f there i s a d i f f e r e n t c o r r e l a t i o n value between cost and s e v e r i t y . While i t would be meaningful to change the sample population to another group of p a t i e n t s , such as non-trauma p a t i e n t s , i t may be quite d i f f i c u l t to apply the i n j u r y s e v e r i t y scale because t h i s scale has been geared to the trauma v i c t i m to measure s e v e r i t y of the l i f e - t h r e a t e n i n g value of the i n d i v i d u a l ' s i n j u r i e s . For example, the s e v e r i t y of a coronary patient or an orthopaedic surgery p a t i e n t may be d i f f i c u l t to measure i n a comparable manner to the s e v e r i t y measurement used i n t h i s research. Other simpler s c a l e s , such as the A.I.S., could be adapted f o r non-trauma victims because the majority of these i l l n e s s e s a f f e c t only one part of the body at one time. There i s no i n d i c a t i o n i n the a n a l y s i s of the r e l a t i o n s h i p between cost and s e v e r i t y for MVA victims to i n d i c a t e the outcome of comparable research f o r non-MVA pa t i e n t s . - 116 -Injury Severity Scale - I.S.S. This scale proved to be more responsive than A.I.S. i n t h i s research, and proved r e l i a b l e i n s e v e r a l other t e s t s . Although the I.S.S. coding system has had l i m i t e d use, due to i t s recent development i n the l a s t f i v e years, i t does seem to have the i n t e r n a l consistency necessary to make v a l i d comparisons. The l i t e r a t u r e review i d e n t i f i e d Semmlow's research which tested I.S.S. and p o t e n t i a l to survive.''" In another study, 2 Detmer studied the r e l a t i o n s h i p of s e v e r i t y and the p o t e n t i a l f o r r e -c e i v i n g poor q u a l i t y of care. Both studies accepted the i n t e r n a l v a r i a t i o n of I.S.S. as being minimal. The r e s u l t s of Detmer's research suggested that the I.S.S. coding system could adequately i d e n t i f y s e v e r i t y and arrange patients i n order of t h e i r s e v e r i t y . Recent studies i n analysis of trauma i n j u r i e s or trauma r e g i s t r y systems show that I.S.S. has become a standard patient c l a s s i f i c a t i o n system for 3 s e v e r i t y . I.S.S. has become p a r t i c u l a r l y valuable when studies are made across d i f f e r e n t h o s p i t a l s e t t i n g s . I t i s u n l i k e l y that an improve-ment i n the c o r r e l a t i o n between the cost and s e v e r i t y would be achieved with another measure of s e v e r i t y . Semmlow J.L., o p . c i t . Detmer D., o p . c i t . lWest J . , Trunkey D., and Lim R., "Systems of Trauma Care - A Study of Two Counties," Archives of Surgery, Vol.114, A p r i l 1979. - 117 -Other Independent Variables Death as an outcome, and whether the patient had an operation, were two independent v a r i a b l e s that seem to have had s i g n i f i c a n t i n f l u e n c e on the r e l a t i o n s h i p between cost and s e v e r i t y . Other v a r i a b l e s could be entered i n t o the equation, to t r y to eliminate some of the i n t e r n a l v a r i a t i o n of cost. The o b j e c t i v e of introducing other independent v a r i -ables i s to reduce the n o n - c l i n i c a l or non-patient care influences on h o s p i t a l episodic costs. As an example, length of stay has been shown to be a c r i t i c a l component of cost a n a l y s i s . Therefore, f a c t o r s which influence length of stay could be considered as valuable independent v a r i a b l e s i n the regression of h o s p i t a l episodic costs and s e v e r i t y . Representative f a c t o r s that influence length of stay include p o t e n t i a l f o r discharge, which may be measured i n a v a i l a b i l i t y of home care or family support, and date of admission, suggesting that weekday admissions d i f f e r from weekend admissions f o r episodic costs when other f a c t o r s are constant. This l a t t e r f a c t o r , date of admission, could be obtained from patients' medical records, but research to determine p o t e n t i a l f o r d i s -charge would be based on interviews with the patient and a l l i e d health workers (such as nursing s t a f f and s o c i a l services personnel), and would therefore involve a large number of assumptions. Step-Down Costing Improving the v a l i d i t y of the component costs that were t o t a l l e d to make Step-Down episodic costs i s the most necessary improvement to the c o r r e l a -t i o n between cost and s e v e r i t y . Determining usable Step-Down episodic costs of a s i n g l e patient stay was an arduous task. Several of the - 118 -methodologies used to develop u n i t costs had the p o t e n t i a l to be r e f i n e d and t h e i r v a l i d i t y increased. While the number of units of s e r v i c e needed by the patient was r e l a t i v e l y easy to c a l c u l a t e f o r each d i v i s i o n of the h o s p i t a l by using p a t i e n t s ' medical records, the d i f f i c u l t i e s arose i n assigning cost to each service u n i t . The most important of these u n i t costs that needs improvement i s nursing costs, as t h i s i s the s i n g l e highest cost component of most pa t i e n t s ' stay i n the h o s p i t a l . I d e n t i f y i n g the i n p a t i e n t beds by s p e c i a l t y , as was done i n t h i s t h e s i s , allowed f o r v a r i a t i o n of s t a f f i n g requirements by nursing u n i t ; i t d i d not, however, i d e n t i f y the i n d i v i d u a l a t t e n t i o n needed by each patient f o r each day of his/her stay. This i d e n t i f i c a t i o n process i s more sop h i s t i c a t e d than assuming that a l l p a t i e n t s ' nursing needs are equiv-alent, but i t i s not as accurate as a patient c l a s s i f i c a t i o n system. An improved i d e n t i f i c a t i o n system would have helped to i d e n t i f y the nursing needs of the patients on many l e v e l s , some of which could have been r e l a t e d to the p a t i e n t ' s i n j u r y , and thus to the s e v e r i t y s c a l e , while others would have been r e l a t e d to d i f f e r e n t personal c h a r a c t e r i s t i c s such as 4 a b i l i t y to communicate, age, and p s y c h o l o g i c a l or s o c i a l needs. Lowell Gerson^ went one step further than simply using patient c l a s s i f i c a t i o n as a management of nursing time by p r e d i c t i n g d a i l y nursing care needs over the p a t i e n t ' s length of stay based on the admitting diagnosis. At the time of h i s research, p r o f i l e s of care were developed f o r j u s t a few ICDA codes Y o u e l l L., "Patient C l a s s i f i c a t i o n Program," Dimension i n Health Service, November 1979, p.17. 'Gersen L., "Patient P r o f i l e s of H o s p i t a l Care," Canadian H o s p i t a l , Vol.50(9), 1973. - 119 -i n the 800 s e r i e s , s p e c i f i c a l l y , those dealing with a c c i d e n t a l i n j u r i e s . Since the A.I.S. code had also been c o r r e l a t e d to the ICDA code, there i s p o t e n t i a l i n future studies to l i n k a s e v e r i t y scale and a patient p r o f i l e s c ale, which would be of great assistance i n d i f f e r e n t i a t i n g the nursing costs of the MVA v i c t i m by s e v e r i t y l e v e l . I f the v a l i d i t y of costs of component services used by each patient could be increased, e s p e c i a l l y f o r nursing and ward costs, the p o t e n t i a l of analyzing the r e l a t i o n s h i p of s e v e r i t y and treatment costs as i d e n t i -f i e d by Cloverdale et a l ^ would be r e a l i z e d . They i d e n t i f i e d the fol l o w i n g three components of h o s p i t a l costs: e Overhead c o s t s — t h o s e costs which vary with the s i z e of the h o s p i t a l rather than i t s throughput; • Hotel c o s t s — t h o s e costs which vary with the number of beds which are occupied i n the h o s p i t a l ; 0 Treatment c o s t s — t h o s e costs which vary with the type and number of patients treated. I f a more r e f i n e d treatment cost could be i d e n t i f i e d , the r e s u l t would b a much c l o s e r r e l a t i o n s h i p with s e v e r i t y . Overhead and h o t e l costs per pat i e n t could then be more e a s i l y predicted, based on the estimated length of stay of the patient as w e l l as s e v e r i t y combined with diagnose Coverdale J . , Gibbs R., and Nurse K., "A Hos p i t a l Cost I'odel f o r P o l i c y A n a l y s i s , " The Journal of the Operational Research Society, Vol.31, Sept. 1980. - 120 -U t i l i t y of Relationship Between Episodic Cost and Severity The u t i l i t y of the r e l a t i o n s h i p between s e v e r i t y and h o s p i t a l episodic costs can be defined by i t s a b i l i t y to i n f l u e n c e h o s p i t a l f i n a n c i a l r e p o r t i n g systems. The greatest s i n g l e f a c t o r i n j u s t i f y i n g and ex-p l a i n i n g h o s p i t a l costs i s patient mix. Assuming that a regression c o - e f f i c i e n t comparable to or better than that produced i n t h i s t h e s i s can be repeated f o r other patient groups, s e v e r i t y has the p o t e n t i a l to become a standard component of patient c l a s s i f i c a t i o n systems which would define patient mix. In the l a s t s e v e r a l years, d i a g n o s t i c - r e l a t e d groups (DRG's) or ICD major codes have been used repeatedly as patient care categories for a wide range of planning issues, i n c l u d i n g models f o r determining bed needs, i n t r o d u c t i o n of new m e d i c a l / s u r g i c a l programs, reimbursement, and p h y s i c i a n recruitment. Rousseau and G i b b J found that ICD chapter headings were too heterogeneous f o r t h e i r needs. To improve t h e i r model f o r simu-l a t i n g h e a l t h care system behavior, they concluded that a d i f f e r e n t c a t e g o r i z a t i o n of patients was needed, such that each category would be more homogeneous with respect to i n t e r n a l e l a s t i c i t i e s (e.g., e l a s t i c i t y of admission r a t e ) . Studies i n v o l v i n g DRG have greater opportunities to reduce i n t e r n a l v a r i a t i o n , as these groups t o t a l 383 compared with j u s t 18 f o r the major ICD codes. But concerns have developed i n using DRG for t h i s purpose, r e s u l t i n g i n a re-organization of p a t i e n t diagnoses i n t o 562 groups. The o r i g i n a l 383 groups were developed f o r u t i l i z a t i o n Rousseau J . , and Gibbs R.J., "A Model to A s s i s t Planning the P r o v i s i o n of H o s p i t a l Services," The Journal of the Operational Research Society, Vol.32, No.6, June 1981. - 121 -review purposes and d i d not s u i t reimbursement s t u d i e s . " The new 562 groups have been defined so as to be c l i n i c a l l y s u i t e d to reimbursement. While t h i s number may appear to be unwieldly f o r general use, an age f a c t o r has been b u i l t i n t o the c h a r a c t e r i z a t i o n of these new groups, to account f o r the v a r i a t i o n i n resource u t i l i z a t i o n of the e l d e r l y ( i . e . , length of stay v a r i a t i o n by age). Most general h o s p i t a l s w i l l use only 40-60 of these new DRG's, as they w i l l not have adequate samples i n the other groups to ensure v a l i d reimbursements f o r care given to t h e i r p a t i e n t s . The i m p l i c a t i o n of data a n a l y s i s of t h i s thesis (that a s e v e r i t y s c a l e -I.S.S. can be responsible f o r 46% of the v a r i a t i o n i n episodic costs under c e r t a i n circumstances), suggests that a new patient c l a s s i f i c a t i o n system needs to be developed which includes a s e v e r i t y component. This new patient c l a s s i f i c a t i o n system would be p a r t i c u l a r l y u s e f u l f o r r e -imbursement an a l y s i s and would have a p p l i c a t i o n i n other long range planning exercises. While the I.S.S. score may be too complicated and not a p p l i c a b l e f o r non-trauma p a t i e n t s , a m o d i f i c a t i o n of the A.I.S. score which measures l i f e - t h r e a t e n i n g diagnoses, could be as e a s i l y included i n a patient c l a s s i f i c a t i o n system as age was included i n the new 562 DRG's. Severity combined with diagnosis provides the v a r i a b l e s f o r an e x c e l l e n t methodology f o r accurately p r e d i c t i n g v a r i a t i o n s i n episodic h o s p i t a l costs. Personal Communication, New Jersey State Cr-partment of Health, Trenton, New Jersey. - 122 -APPENDIX A Implications of Data Analysis i n Reference to I.C.B.C. The sample population was drawn from motor v e h i c l e accident victims f o r reasons l i s t e d i n the main text of t h i s report, and to a s s i s t I.C.B.C. i n i t s long range planning. Two major f i n d i n g s that could be important to I.C.B.C. would be the comparison of Step-Down to Per Diem costs of MVA victims and the percentage of victims i n each c e l l of the s e v e r i t y scale. Comparing Ep i s o d i c Costs As a s i n g l e group of i n p a t i e n t s , MVA victims had an average episodic cost, using step-down costing methodology, of $1,936.10. This was j u s t s l i g h t l y below the average episodic cost, using per diem costing methodology, of $1,984.15. While i n d i v i d u a l patient's costs varied s i g n i f i c a n t l y from one methodology to the other, there was no s i g n i f i -cant d i f f e r e n c e when a l l patients' costs were t o t a l l e d together. This data could be of value i f and when, i n the future, h o s p i t a l reimburse-ments are based on case mix, or a combination of diagnoses and s e v e r i t y . If the system were to change, there i s no i n d i c a t i o n that there would be a s i g n i f i c a n t increase or decrease i n the t o t a l h o s p i t a l cost to t r e a t a l l motor accident victims, compared to the present method of reimbursement using a set per diem r a t e . 122A LITERATURE CITED - 123 -P a t i e n t D i s t r i b u t i o n by Severity The l a r g e s t concentration of MVA i n p a t i e n t s was i n the low s e v e r i t y range, with an I.S.S. value of l e s s than 10. I n d i v i d u a l patients with low s e v e r i t y were shown to have lower h o s p i t a l episodic costs than those p a t i e n t s with higher s e v e r i t y values. But as a group, the com-bined costs of low s e v e r i t y patients were l a r g e r than the combined costs of a l l other p a t i e n t s . If future preventive programs are to be focussed on reducing deaths on B r i t i s h Columbia highways, there would not l i k e l y be a reduction i n health care costs. This statement i s made on the assumption that a large percentage of those who survive w i l l have a high s e v e r i t y r a t i n g , with a correspondingly high h o s p i t a l cost f o r acute care. ( A d d i t i o n a l costs to the health care system would also be incurred f o r extended, convales-cent and r e h a b i l i t a t i o n care.) If future preventive programs are to be focussed on reducing s e v e r i t y of a l l i n j u r i e s i n equal proportion, then reductions i n h o s p i t a l cost could be r e a l i z e d . For example, i f a 10% s h i f t downward were r e a l i z e d f o r each of the f i v e c e l l s of A.I.S., there would not be an o v e r a l l reduction of 10% of the h o s p i t a l s ' patients (see following graph); but there would be a s h i f t to the low s e v e r i t y groups, e s p e c i a l l y A . I . S . - l and A.I.S.-2, which have lower h o s p i t a l episodic costs. - 124 -906 H o s p i t a l i z e d MVA i n 1975 i n Vancouver General H o s p i t a l A.I.S. Level Number of Patients 10% Reduction New Tot a l 5 29 ( 29- 3)= 26 26 + 10% of Deaths 4 97 ( 97-10)= 87 87 + 3= 90 3 610 (610-61)=549 549 + 10=559 2 143 (143-14)=129 129 + 61=190 1 27 ( 27- 3)= 24 24 + 14= 38 - 125 -LITERATURE CITED American H o s p i t a l A s s o c i a t i o n , Cost Findings f o r H o s p i t a l s , Chicago, 1957. Baker S. et a l , "The Injury Severity Score - A Method for Describing Patients with M u l t i p l e I n j u r i e s and Evaluating Emergency Care," Journal of Trauma, March 1974. Baker S.P., O ' N e i l l B., Haddon W., and Long W.B., "The Injury Severity Scale: Development and P o t e n t i a l Usefulness" Proceedings of the 18th Conference of the American A s s o c i a t i o n f o r Automobile Medicine. Lake B l u f f , 111. AAAM 1974. Berry R.E., "Product Heterogenity and H o s p i t a l Cost A n a l y s i s , " Inquiry, 7-67, March 1970. Bowman R., "DRG's Help Trace Revenue Sources," Modern Health Care, January 1980. Boyd D.R. et a l , "Trauma Registry New Computer Method for M u l t i f a c t o r i a l Evaluation of Major Health Problems," JAMA Vol.223,1973. Chipman M.L., " H o s p i t a l i z a t i o n a f t e r MVA i n Nova S c o t i a , " Canadian Journal of P u b l i c Health, Vol.64, March 1973. Committee on Medical Aspects of Automotive Safety, "Rating the Severity of Tissue Damage I The Abbreviated Injury Scale," JAMA, Jan 11 1971, Vol.215 No. 2. Committee on Medical Aspects of Automotive Safety, "Rating the Severity of Tissue Damage II The Comprehensive Injury Scale," May 1 1972, Vol.220 No. 5. Coverdale J . , Gibbs R., and Nurse K., "A H o s p i t a l Cost Model f o r P o l i c y A n a l y s i s , " The Journal of the Operational Research Society, Vol.31, Sept.1980. Cowley et a l , "A Prognostic Index f o r Severe Trauma," Journal of Trauma, December 1974. Decoster D.T., Schafer E.L., Management Accounting, Wiley and Sons, Toronto, 1979, pg.693. Detmer D., et a l , "Quality of Care for Major Trauma," Journal of Trauma, Vol.16, No.7, 1976, pg.522. Dittman D.A., and Smith K.R., "Consideration of Benefits and Costs: A Conceptual Framework f o r the Health Planner," Health Care Management  Review, F a l l 1979. E l n i c k i R.A., "H o s p i t a l P r o d u c t i v i t y , Service I n t e n s i t y , and Cost," Health Services Research, Winter 1974. Evans R.G., and Robinson G.C., "Evaluation of the Economic Implications of a Day Care Surgery Unit," Report to Children's H o s p i t a l , Vancouver, B.C., 1973. - 126 -Fargan B.M., "1975 S o c i e t a l Costs of Motor Vehicle Accidents," U.S. Department of Transportation, National Highway T r a f f i c Safety Administration, December 1976, page 30. F e l d s t e i n M., " H o s p i t a l Cost V a r i a t i o n and Case Mix Differences," Medical Care, 3, 95-103, April-June 1965. Frank W.G., "A Managerial Accounting Analysis of H o s p i t a l Costs," Health Services Research,. Spring 1976. Gerson L., "Patient P r o f i l e s of H o p s i t a l Care," Canadian H o s p i t a l , Vol.50(9), 1973. Grosse R.N., "Cost-Benefit Analysis of Health Services," American  Academy of P o l i t i c a l and S o c i a l Sciences, P h i l . Annuls 399, 1972. Haddon W., "Reducing Damage of Motor Vehicle Use," Technology Review, July/August, 1975. Haddon W., Suchman E.A., and K l e i n D., Accident Research - Methods and  Approaches, New York, Harper & Row, 1964. Haddon W., "The T r a n s i t i o n to Approaches E t i o l o g i c a l l y rather than D e s c r i p t i v e l y Based, with respect to Epidemiology Prevention and Amelioration of Trauma," American Journal of P u b l i c Health, Vol.58, August 1968. Holder W.W., " H o s p i t a l Bedgeting: State of the Ar t , " Hospitals and Health  Services Administration, Vol.23, No.2, Spring 1978. Klarman H.E., "Applications of Cost Benefit Analysis to the Health Services and the S p e c i a l Case of Technologic Innovation," Int ernat i o n a l  Journal of Health Services, Vol.4, No.2, Nov 2, 1974. Krischer J.P., "Indexes of Severity: Underlying Concepts," Health  Services Research, Summer 1976. Lalonde M., "A New Perspective of the Health of Canadians: A Working Document," Ottawa, A p r i l 1974. Lave J.R., and Lave L.B., "H o s p i t a l Cost Functions," American Economic  Review, 60:379, June 1970. Lave J.R., and Lave L.B., "The Extent of Role D i f f e r e n t i a t i o n Among Hos p i t a l s , " Health Services Research, Spring 1971. Lay C , "What i s Disease Costing?" Unpublished Report, Health Admin-i s t r a t i o n , Faculty of Administration, U n i v e r s i t y of Ottawa. Lynch T., "Report to the Information Systems Steering Committee of Ontario M i n i s t r y of Health," Ontario H o s p i t a l A s s o c i a t i o n , October 20, 1978. Macdonald L.K., and Reuter F., "A Patient S p e c i f i c Approach to H o s p i t a l Cost Accounting," Health Services Research, Summer 1973. - 127 -Magraw R.W., M.D., "How Trends i n Medical Education are A f f e c t i n g Medicine and H o s p i t a l , " Hospitals, October 1, 1963. Martin D.L., Health Program Evaluation - A Primer Report to the F e d e r a l - P r o v i n c i a l Sub-Committee on Quality of Care and Research and Fe d e r a l - P r o v i n c i a l Advisory Committee on Health Insurance, 1977 Vol.1 Martin D.L., Health Program Evaluation - A Primer Report to the F e d e r a l - P r o v i n c i a l Advisory Committee on Health Insurance, 1977 Vol.11 Moylan M.D. et a l , "Evaluation of the Quali t y of H o s p i t a l Care f o r Major Trauma," Journal of Trauma, Vol.16 #7, J u l y 1976. Murray J.E., " E t i o l o g y of MVA with S p e c i a l Reference to the Mechanism of Injury," N.E. Journal of Medicine NE 492, June 20, 1968. Musgrave R.A., "Cost Benefit Analysis and the Theory of P u b l i c Finance," Journal of Economic L i t e r a t u r e , Vol.7, No.3, 1969. Mushkin S.J. and C o l l i n g s F.d'A., "Economic Costs of Disease and Injury," P u b l i c Health Reports, Vol.74, No.9, Sept. 1959. Noble J . et a l , Emergency Medical Services: Behavioural Planning  Perspectives, New York, Behavioural P u b l i c a t i o n s , 1973. Ogawa M. and Sugimoto T., "Rating Severity of the Injured by Ambulance Attendant; F i e l d Research of Trauma Index," Journal of Trauma, Nov. 1974. Rice D.P., "Estimating the Cost of I l l n e s s , " American Journal of Publ i c  Health, Vol.57, No.3, March 1967. Rousseau J . , and Gibbs R.J., "A Model to A s s i s t Planning the Pr o v i s i o n of H o s p i t a l Services," The Journal of the Operational Research Society, Vol.32, No.6, June 1981. Schwenger C , "Alcohol and Highway Safety," Report to Congress from the Secretary of Transportation, 1967. Semmlow J.L. and Cone R., " U t i l i t y of the Injury Severity Score: A Confirmation," Health Services Research, Spring 1976. Stanford Center f o r Health Care Research, "Comparison of Hospitals with Regard to Outcomes of Surgery," Health Services Research, Summer 1976. Stoddart G.L., On Determining the E f f i c i e n c y of Health Programs, Unpublished a r t i c l e , Department of C l i n i c a l Epidemology and B i o s t a t i s t i c s and Department of Economics, McMaster U n i v e r s i t y , November 1980. Thompson J.D., "On Reasonable Cost of H o s p i t a l Services," Millbank  Memorial Fund Quarterly, 46(1):33 January 1968 Part 2. - 128 -Thompson J.D., A v e r i l l R.F. , and Fe t t e r R.B., "Planning Budgeting and Controlling-One Look at the Future: Case Mix Cost Accounting," Health  Services Research, Summer 1979. Torrance G.W., Thomas W.H., and Sackett D.L., "A U t i l i t y Maximization Model f o r Evaluation of Health Care Programs," Health Services Research, Summer 1972. Weinstein M.C., and Stason W.B., " A l l o c a t i o n of Resources to Manage Hypertension," The New England Journal of Medicine, March 31, 1977. Weinstein M.C., and Stason W.B., "Foundations of Cost-Effectiveness An a l y s i s f o r Health and Medical P r a c t i c e s , " The New England Journal of  Medicine, March 31, 1977 West J.,'Trunkey D., and Lim R., "Systems of Trauma Care - A Study of Two Counties," Archives of Surgery, Vol.114, A p r i l 1979. W.H.O., "Accidents i n a S o c i a l Context," Chronicle 27, 1973. Youel l L., "Patient C l a s s i f i c a t i o n Program," Dimensions i n Health Service, November 1979, p.17. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0095314/manifest

Comment

Related Items